Daily Papers

Heterogeneous LLM serving stacks split scheduling into two layers that optimize in isolation: model routers pick a model from quality and cost signals while ignoring instance load, and serving load balancers optimize queues while ignoring quality. We present RouteBalance, a serving-aware scheduling layer that fuses both into a single online assignment over concrete model instances, jointly trading off quality, latency, and cost. A batched in-process predictor stack and dead-reckoned instance state keep the joint decision cheap on the request hot path (approx32 ms at 12 req/s). On a 13-instance, 28-GPU heterogeneous cluster serving four model sizes, a single deployed RouteBalance stack traces the upper region of the three-way quality-cost-throughput frontier. Sweeping one weight vector reaches both the highest routing-decision quality (DeepEval 0.419, +0.013 over the strongest baseline, 95% CI [{+}0.005,{+}0.022]; the ordering holds when a second judge re-scores the actually served text) and, at its cost-priority corner, per-request cost that ties the cheapest baseline. With router engineering equalized against concurrent-scoring baseline variants we build, its balanced preset serves at 2.8 s and 30 req/s, leading 2.6 to 4.1times ahead of enhanced BEST-Route at high load. (Deploying those routers as published, one serial scoring call per request, makes them collapse 23times under load, a deployment-architecture effect we isolate separately, not the routing result.) A four-arm isolation shows the benefit follows from pricing latency at model-selection time; the learned predictors contribute calibration and SLO headroom rather than the headline frontier. Code: https://github.com/AKafakA/route-balance

Interest in neural machine translation has grown rapidly as its effectiveness has been demonstrated across language and data scenarios. New research regularly introduces architectural and algorithmic improvements that lead to significant gains over "vanilla" NMT implementations. However, these new techniques are rarely evaluated in the context of previously published techniques, specifically those that are widely used in state-of-theart production and shared-task systems. As a result, it is often difficult to determine whether improvements from research will carry over to systems deployed for real-world use. In this work, we recommend three specific methods that are relatively easy to implement and result in much stronger experimental systems. Beyond reporting significantly higher BLEU scores, we conduct an in-depth analysis of where improvements originate and what inherent weaknesses of basic NMT models are being addressed. We then compare the relative gains afforded by several other techniques proposed in the literature when starting with vanilla systems versus our stronger baselines, showing that experimental conclusions may change depending on the baseline chosen. This indicates that choosing a strong baseline is crucial for reporting reliable experimental results.

DARTS search space (DSS) has become a canonical benchmark for NAS whereas some emerging works pointed out the issue of narrow accuracy range and claimed it would hurt the method ranking. We observe some recent studies already suffer from this issue that overshadows the meaning of scores. In this work, we first propose and orchestrate a suite of improvements to frame a larger and harder DSS, termed LHD, while retaining high efficiency in search. We step forward to renovate a LHD-based new benchmark, taking care of both discernibility and accessibility. Specifically, we re-implement twelve baselines and evaluate them across twelve conditions by combining two underexpolored influential factors: transductive robustness and discretization policy, to reasonably construct a benchmark upon multi-condition evaluation. Considering that the tabular benchmarks are always insufficient to adequately evaluate the methods of neural architecture search (NAS), our work can serve as a crucial basis for the future progress of NAS. https://github.com/chaoji90/LHD

While Large Language Models (LLMs) excel on standardized medical exams, high scores often fail to translate to high-quality responses for real-world medical queries. Current evaluations rely heavily on multiple-choice questions, failing to capture the unstructured, ambiguous, and long-tail complexities inherent in genuine user inquiries. To bridge this gap, we introduce QuarkMedBench, an ecologically valid benchmark tailored for real-world medical LLM assessment. We compiled a massive dataset spanning Clinical Care, Wellness Health, and Professional Inquiry, comprising 20,821 single-turn queries and 3,853 multi-turn sessions. To objectively evaluate open-ended answers, we propose an automated scoring framework that integrates multi-model consensus with evidence-based retrieval to dynamically generate 220,617 fine-grained scoring rubrics (~9.8 per query). During evaluation, hierarchical weighting and safety constraints structurally quantify medical accuracy, key-point coverage, and risk interception, effectively mitigating the high costs and subjectivity of human grading. Experimental results demonstrate that the generated rubrics achieve a 91.8% concordance rate with clinical expert blind audits, establishing highly dependable medical reliability. Crucially, baseline evaluations on this benchmark reveal significant performance disparities among state-of-the-art models when navigating real-world clinical nuances, highlighting the limitations of conventional exam-based metrics. Ultimately, QuarkMedBench establishes a rigorous, reproducible yardstick for measuring LLM performance on complex health issues, while its framework inherently supports dynamic knowledge updates to prevent benchmark obsolescence.

Behaviour-Driven Development (BDD) suites accumulate step-text duplication whose maintenance cost is established in prior work. Existing detection techniques require running the tests (Binamungu et al., 2018-2023) or are confined to a single organisation (Irshad et al., 2020-2022), leaving a gap: a purely static, paraphrase-robust, step-level detector usable on any repository. We fill the gap with cukereuse, an open-source Python CLI combining exact hashing, Levenshtein ratio, and sentence-transformer embeddings in a layered pipeline, released alongside an empirical corpus of 347 public GitHub repositories, 23,667 parsed .feature files, and 1,113,616 Gherkin steps. The step-weighted exact-duplicate rate is 80.2 %; the median-repository rate is 58.6 % (Spearman rho = 0.51 with size). The top hybrid cluster groups 20.7k occurrences across 2.2k files. Against 1,020 pairs manually labelled by the three authors under a released rubric (inter-annotator Fleiss' kappa = 0.84 on a 60-pair overlap), we report precision, recall, and F1 with bootstrap 95 % CIs under two protocols: the primary rubric and a score-free second-pass relabelling. The strongest honest pair-level number is near-exact at F1 = 0.822 on score-free labels; the primary-rubric semantic F1 = 0.906 is inflated by a stratification artefact that pins recall at 1.000. Lexical baselines (SourcererCC-style, NiCad-style) reach primary F1 = 0.761 and 0.799. The paper also presents a CDN-structured critique of Gherkin (Cognitive Dimensions of Notations); eight of fourteen dimensions are rated problematic or unsupported. The tool, corpus, labelled pairs, rubric, and pipeline are released under permissive licences.

Training algorithms, broadly construed, are an essential part of every deep learning pipeline. Training algorithm improvements that speed up training across a wide variety of workloads (e.g., better update rules, tuning protocols, learning rate schedules, or data selection schemes) could save time, save computational resources, and lead to better, more accurate, models. Unfortunately, as a community, we are currently unable to reliably identify training algorithm improvements, or even determine the state-of-the-art training algorithm. In this work, using concrete experiments, we argue that real progress in speeding up training requires new benchmarks that resolve three basic challenges faced by empirical comparisons of training algorithms: (1) how to decide when training is complete and precisely measure training time, (2) how to handle the sensitivity of measurements to exact workload details, and (3) how to fairly compare algorithms that require hyperparameter tuning. In order to address these challenges, we introduce a new, competitive, time-to-result benchmark using multiple workloads running on fixed hardware, the AlgoPerf: Training Algorithms benchmark. Our benchmark includes a set of workload variants that make it possible to detect benchmark submissions that are more robust to workload changes than current widely-used methods. Finally, we evaluate baseline submissions constructed using various optimizers that represent current practice, as well as other optimizers that have recently received attention in the literature. These baseline results collectively demonstrate the feasibility of our benchmark, show that non-trivial gaps between methods exist, and set a provisional state-of-the-art for future benchmark submissions to try and surpass.

The rapid proliferation of Large Language Models (LLMs) and diverse specialized benchmarks necessitates a shift from fragmented, task-specific metrics to a holistic, competitive ranking system that effectively aggregates performance across multiple ability dimensions. Primarily using static scoring, current evaluation methods are fundamentally limited. They struggle to determine the proper mix ratio across diverse benchmarks, and critically, they fail to capture a model's dynamic competitive fitness or its vulnerability when confronted with sequential, high-stakes tasks. To address this, we introduce the novel Competitive Swiss-System Dynamics (CSD) framework. CSD simulates a multi-round, sequential contest where models are dynamically paired across a curated sequence of benchmarks based on their accumulated win-loss record. And Monte Carlo Simulation (N=100,000 iterations) is used to approximate the statistically robust Expected Win Score (E[S_m]), which eliminates the noise of random pairing and early-round luck. Furthermore, we implement a Failure Sensitivity Analysis by parameterizing the per-round elimination quantity (T_k), which allows us to profile models based on their risk appetite--distinguishing between robust generalists and aggressive specialists. We demonstrate that CSD provides a more nuanced and context-aware ranking than traditional aggregate scoring and static pairwise models, representing a vital step towards risk-informed, next-generation LLM evaluation.

Recent advances in large language model (LLM) embeddings have enabled powerful representations for biological data, but most applications to date focus only on gene-level information. We present one of the first systematic frameworks to generate variant-level embeddings across the entire human genome. Using curated annotations from FAVOR, ClinVar, and the GWAS Catalog, we constructed semantic text descriptions for 8.9 billion possible variants and generated embeddings at three scales: 1.5 million HapMap3+MEGA variants, ~90 million imputed UK Biobank variants, and ~9 billion all possible variants. Embeddings were produced with both OpenAI's text-embedding-3-large and the open-source Qwen3-Embedding-0.6B models. Baseline experiments demonstrate high predictive accuracy for variant properties, validating the embeddings as structured representations of genomic variation. We outline two downstream applications: embedding-informed hypothesis testing by extending the Frequentist And Bayesian framework to genome-wide association studies, and embedding-augmented genetic risk prediction that enhances standard polygenic risk scores. These resources, publicly available on Hugging Face, provide a foundation for advancing large-scale genomic discovery and precision medicine.

The growing awareness of safety concerns in large language models (LLMs) has sparked considerable interest in the evaluation of safety within current research endeavors. This study investigates an interesting issue pertaining to the evaluation of LLMs, namely the substantial discrepancy in performance between multiple-choice questions and open-ended questions. Inspired by research on jailbreak attack patterns, we argue this is caused by mismatched generalization. That is, the LLM does not have a comprehensive understanding of the complex concept of safety. Instead, it only remembers what to answer for open-ended safety questions, which makes it unable to solve other forms of safety tests. We refer to this phenomenon as fake alignment and construct a comparative benchmark to empirically verify its existence in LLMs. Such fake alignment renders previous evaluation protocols unreliable. To address this, we introduce the Fake alIgNment Evaluation (FINE) framework and two novel metrics--Consistency Score (CS) and Consistent Safety Score (CSS), which jointly assess two complementary forms of evaluation to quantify fake alignment and obtain corrected performance estimates. Applying FINE to 14 widely-used LLMs reveals several models with purported safety are poorly aligned in practice. Our work highlights potential limitations in prevailing alignment methodologies.

In this position paper, we argue that human baselines in foundation model evaluations must be more rigorous and more transparent to enable meaningful comparisons of human vs. AI performance, and we provide recommendations and a reporting checklist towards this end. Human performance baselines are vital for the machine learning community, downstream users, and policymakers to interpret AI evaluations. Models are often claimed to achieve "super-human" performance, but existing baselining methods are neither sufficiently rigorous nor sufficiently well-documented to robustly measure and assess performance differences. Based on a meta-review of the measurement theory and AI evaluation literatures, we derive a framework with recommendations for designing, executing, and reporting human baselines. We synthesize our recommendations into a checklist that we use to systematically review 115 human baselines (studies) in foundation model evaluations and thus identify shortcomings in existing baselining methods; our checklist can also assist researchers in conducting human baselines and reporting results. We hope our work can advance more rigorous AI evaluation practices that can better serve both the research community and policymakers. Data is available at: https://github.com/kevinlwei/human-baselines

We conduct a systematic audit of three widely used reasoning benchmarks, SocialIQa, FauxPas-EAI, and ToMi, and uncover pervasive flaws in both benchmark items and evaluation methodology. Using five LLMs (GPT-{3, 3.5, 4, o1}, and LLaMA 3.1) as diagnostic tools, we identify structural, semantic, and pragmatic issues in benchmark design (e.g., duplicated items, ambiguous wording, and implausible answers), as well as scoring procedures that prioritize output form over reasoning process. Through systematic human annotation and re-evaluation on cleaned benchmark subsets, we find that model scores often improve not due to due to erratic surface wording variations and not to improved reasoning. Infact, further analyses show that model performance is highly sensitive to minor input variations such as context availability and phrasing, revealing that high scores may reflect alignment with format-specific cues rather than consistent inference based on the input. These findings challenge the validity of current benchmark-based claims about reasoning in LLMs, and highlight the need for evaluation protocols that assess reasoning as a process of drawing inference from available information, rather than as static output selection. We release audited data and evaluation tools to support more interpretable and diagnostic assessments of model reasoning.

Autonomous research agents produce competitive solutions and professional-looking manuscripts, yet their outputs contain verifiability failures undetectable by surface-level evaluation: fabricated citations, unreproducible scores, and method descriptions that diverge from the implementation. We address this through three contributions. First, Chain-of-Evidence (CoE), a verifiability framework requiring every claim to be traceable to its evidence source. Second, ScientistOne, an end-to-end autonomous research system that maintains evidence chains by construction throughout literature review, solution discovery, and paper writing. Third, CoE Audit, a post-hoc audit whose four integrity checks -- score verification, specification violation, reference verification, and method-code alignment -- apply uniformly to all systems. Across 75 papers spanning five systems and five frontier research tasks, every baseline exhibits at least one systematic failure mode: hallucinated reference rates reach 21%, score verification passes in as few as 42% of papers, and method-code alignment ranges from 20% to 80%. ScientistOne achieves zero hallucinated references (0/337), perfect score verification (12/12), and the highest method-code alignment (14/15), while matching or exceeding human expert performance on all five tasks. ScientistOne further generalizes to six additional tasks spanning medical imaging, fine-grained recognition, 3D perception, and language modeling, achieving state-of-the-art on Parameter Golf and gold medals on MLE-Bench tasks where baselines fail entirely.

Reinforcement Learning (RL) for Large Language Models (LLMs) often suffers from training collapse in long-horizon tasks due to exploding gradient variance. To mitigate this, a baseline is commonly introduced for advantage computation; however, traditional value models remain difficult to optimize, and standard group-based baselines overlook sequence heterogeneity. Although classic optimal baseline theory can achieve global variance reduction, it neglects token heterogeneity and requires prohibitive gradient-based computation. In this work, we derive the Optimal Token Baseline (OTB) from first principles, proving that gradient updates should be weighted inversely to their cumulative gradient norm. To ensure efficiency, we propose the Logit-Gradient Proxy that approximates the gradient norm using only forward-pass probabilities. Our method achieves training stability and matches the performance of large group sizes (N=32) with only N=4, reducing token consumption by over 65% across single-turn and tool-integrated reasoning tasks.

The deployment of machine learning models in operational contexts represents a significant investment for any organisation. Consequently, the risk of these models being misappropriated by competitors needs to be addressed. In recent years, numerous proposals have been put forth to detect instances of model stealing. However, these proposals operate under implicit and disparate data and model access assumptions; as a consequence, it remains unclear how they can be effectively compared to one another. Our evaluation shows that a simple baseline that we introduce performs on par with existing state-of-the-art fingerprints, which, on the other hand, are much more complex. To uncover the reasons behind this intriguing result, this paper introduces a systematic approach to both the creation of model fingerprinting schemes and their evaluation benchmarks. By dividing model fingerprinting into three core components -- Query, Representation and Detection (QuRD) -- we are able to identify sim100 previously unexplored QuRD combinations and gain insights into their performance. Finally, we introduce a set of metrics to compare and guide the creation of more representative model stealing detection benchmarks. Our approach reveals the need for more challenging benchmarks and a sound comparison with baselines. To foster the creation of new fingerprinting schemes and benchmarks, we open-source our fingerprinting toolbox.

Machine learning in high-stakes domains such as healthcare requires not only strong predictive performance but also reliable uncertainty quantification (UQ) to support human oversight. Multi-label text classification (MLTC) is a central task in this domain, yet remains challenging due to label imbalances, dependencies, and combinatorial complexity. Existing MLTC benchmarks are increasingly saturated and may be affected by training data contamination, making it difficult to distinguish genuine reasoning capabilities from memorization. We introduce MADE, a living MLTC benchmark derived from {m}edical device {ad}verse {e}vent reports and continuously updated with newly published reports to prevent contamination. MADE features a long-tailed distribution of hierarchical labels and enables reproducible evaluation with strict temporal splits. We establish baselines across more than 20 encoder- and decoder-only models under fine-tuning and few-shot settings (instruction-tuned/reasoning variants, local/API-accessible). We systematically assess entropy-/consistency-based and self-verbalized UQ methods. Results show clear trade-offs: smaller discriminatively fine-tuned decoders achieve the strongest head-to-tail accuracy while maintaining competitive UQ; generative fine-tuning delivers the most reliable UQ; large reasoning models improve performance on rare labels yet exhibit surprisingly weak UQ; and self-verbalized confidence is not a reliable proxy for uncertainty. Our work is publicly available at https://hhi.fraunhofer.de/aml-demonstrator/made-benchmark.

Many tasks revolve around editing a document, whether code or text. We formulate the revision similarity problem to unify a wide range of machine learning evaluation problems whose goal is to assess a revision to an existing document. We observe that revisions usually change only a small portion of an existing document, so the existing document and its immediate revisions share a majority of their content. We formulate five adequacy criteria for revision similarity measures, designed to align them with human judgement. We show that popular pairwise measures, like BLEU, fail to meet these criteria, because their scores are dominated by the shared content. They report high similarity between two revisions when humans would assess them as quite different. This is a fundamental flaw we address. We propose a novel static measure, Excision Score (ES), which computes longest common subsequence (LCS) to remove content shared by an existing document with the ground truth and predicted revisions, before comparing only the remaining divergent regions. This is analogous to a surgeon creating a sterile field to focus on the work area. We use approximation to speed the standard cubic LCS computation to quadratic. In code-editing evaluation, where static measures are often used as a cheap proxy for passing tests, we demonstrate that ES surpasses existing measures. When aligned with test execution on HumanEvalFix, ES improves over its nearest competitor, SARI, by 12% Pearson correlation and by >21% over standard measures like BLEU. The key criterion is invariance to shared context; when we perturb HumanEvalFix with increased shared context, ES' improvement over SARI increases to 20% and >30% over standard measures. ES also handles other corner cases that other measures do not, such as correctly aligning moved code blocks, and appropriately rewarding matching insertions or deletions.

We introduce CAT, a framework designed to evaluate and visualize the interplay of accuracy and response consistency of Large Language Models (LLMs) under controllable input variations, using multiple-choice (MC) benchmarks as a case study. Current evaluation practices primarily focus on model capabilities such as accuracy or benchmark scores and, more recently, measuring consistency is being considered an essential property for deploying LLMs in high-stake, real-world applications. We argue in this paper that although both dimensions should still be evaluated independently, their inter-dependency also need to be considered for a more nuanced evaluation of LLMs. At the core of CAT are the Consistency-Accuracy Relation (CAR) curves, which visualize how model accuracy varies with increasing consistency requirements, as defined by the Minimum-Consistency Accuracy (MCA) metric. We further propose the Consistency-Oriented Robustness Estimate (CORE) index, a global metric that combines the area and shape of the CAR curve to quantify the trade-off between accuracy and consistency. We present a practical demonstration of our framework across a diverse set of generalist and domain-specific LLMs, evaluated on multiple MC benchmarks. We also outline how CAT can be extended beyond MC tasks to support long-form, open-ended evaluations through adaptable scoring functions.

Common evaluation paradigms for language models focus on scoring single responses through accuracy metrics or proper scoring rules, failing to capture the full richness of a model's belief state. Recent work illustrates that language models hallucinate in-part because they are optimised to be good test-takers under binary scoring schemes that reward any answer over abstention. While this insight naturally leads to penalty-based approaches, they ignore crucial distinctions in how models distribute uncertainty, for example between hedging toward incorrect answers versus hedging toward "I don't know" responses. A novel evaluation metric, the Distributional Correctness Score (DCS), is introduced to solve this problem, i.e., of not considering a model's entire probability distribution over answer choices. DCS naturally distinguishes between harmful overconfidence in wrong answers and uncertainty expressed through abstention, providing scores in an interpretable default range. Through theoretical analysis and illustrative examples, DCS is demonstrated to offer a more nuanced and aligned evaluation paradigm that incentivises models to express genuine uncertainty rather than guessing. Adapting 12 existing evaluation benchmarks to DCS's variants and measuring performance on six language models reveals that for half of the tested benchmarks scores are negative across all tested models, indicating significant tendencies towards hallucination.

Context. Behaviour-Driven Development (BDD) software test suites accumulate duplicated step subsequences. Three published refactoring patterns are available (within-file Background, within-repo reusable-scenario invocation, cross-organisational shared higher-level step), but no prior work automates which recurring subsequences are worth extracting or which mechanism applies. Objective. Rank recurring step subsequences ("slices") by refactoring suitability (extraction-worthy), pre-map each to one of the three patterns, and quantify prevalence across the public BDD ecosystem. Method. Every contiguous L-step window (L in [2, 18]) in a 339-repository / 276-upstream-owner Gherkin corpus is keyed by paraphrase-robust cluster identifiers and counted under three scopes. Sentence-BERT (SBERT) / Uniform Manifold Approximation and Projection (UMAP) / Hierarchical Density-Based Clustering (HDBSCAN) recovers paraphrase-equivalent slices. Three authors label a stratified 200-slice pool against a written rubric. An eXtreme Gradient Boosting (XGBoost) extraction-worthy classifier trained under 5-fold cross-validation is compared with a tuned rule baseline and two open-weight Large Language Model (LLM) judges. Results. The miner produces 5,382,249 slices collapsing to 692,020 recurring patterns. Three-author Fleiss' kappa = 0.56 (extraction-worthy) and 0.79 (mechanism). The classifier reaches out-of-fold F1 = 0.891 (95% CI [0.852, 0.927]), outperforming both the rule baseline (F1 = 0.836, p = 0.017) and the better LLM judge (F1 = 0.728, p < 1e-4). 75.0%, 59.5%, and 11.7% of scenarios carry a within-file Background, within-repo reusable-scenario, or cross-organisational shared-step candidate. Conclusion. Paraphrase-robust subscenario discovery yields a corpus-wide census of BDD refactoring opportunities; pipeline, classifier predictions, labelled pool, and rubric are released under Apache-2.0.

We revisit test-time scaling for language model reasoning and ask a fundamental question: at equal token budget and compute, is it better to run multiple independent chains in parallel, or to run fewer chains that iteratively refine through sequential steps? Through comprehensive evaluation across 5 state-of-the-art open source models and 3 challenging reasoning benchmarks, we find that sequential scaling where chains explicitly build upon previous attempts consistently outperforms the dominant parallel self-consistency paradigm in 95.6% of configurations with gains in accuracy upto 46.7%. Further, we introduce inverse-entropy weighted voting, a novel training-free method to further boost the accuracy of sequential scaling. By weighing answers in proportion to the inverse entropy of their reasoning chains, we increase our success rate over parallel majority and establish it as the optimal test-time scaling strategy. Our findings fundamentally challenge the parallel reasoning orthodoxy that has dominated test-time scaling since Wang et al.'s self-consistency decoding (Wang et al., 2022), positioning sequential refinement as the robust default for modern LLM reasoning and necessitating a paradigm shift in how we approach inference-time optimization.

The evaluation of large language models (LLMs) via benchmarks is widespread, yet inconsistencies between different leaderboards and poor separability among top models raise concerns about their ability to accurately reflect authentic model capabilities. This paper provides a critical analysis of benchmark effectiveness, examining main-stream prominent LLM benchmarks using results from diverse models. We first propose a new framework for accurate and reliable estimations of item characteristics and model abilities. Specifically, we propose Pseudo-Siamese Network for Item Response Theory (PSN-IRT), an enhanced Item Response Theory framework that incorporates a rich set of item parameters within an IRT-grounded architecture. Based on PSN-IRT, we conduct extensive analysis which reveals significant and varied shortcomings in the measurement quality of current benchmarks. Furthermore, we demonstrate that leveraging PSN-IRT is able to construct smaller benchmarks while maintaining stronger alignment with human preference.

Benchmarks are pivotal in driving AI progress, and invalid benchmark questions frequently undermine their reliability. Manually identifying and correcting errors among thousands of benchmark questions is not only infeasible but also a critical bottleneck for reliable evaluation. In this work, we introduce a framework for systematic benchmark revision that leverages statistical analysis of response patterns to flag potentially invalid questions for further expert review. Our approach builds on a core assumption commonly used in AI evaluations that the mean score sufficiently summarizes model performance. This implies a unidimensional latent construct underlying the measurement experiment, yielding expected ranges for various statistics for each item. When empirically estimated values for these statistics fall outside the expected range for an item, the item is more likely to be problematic. Across nine widely used benchmarks, our method guides expert review to identify problematic questions with up to 84\% precision. In addition, we introduce an LLM-judge first pass to review questions, further reducing human effort. Together, these components provide an efficient and scalable framework for systematic benchmark revision.

We present AutoBench, a fully automated and self-sustaining framework for evaluating Large Language Models (LLMs) through reciprocal peer assessment. This paper provides a rigorous scientific validation of the AutoBench methodology, originally developed as an open-source project by eZecute S.R.L.. Unlike static benchmarks that suffer from test-set contamination and limited adaptability, AutoBench dynamically generates novel evaluation tasks while models alternately serve as question generators, contestants, and judges across diverse domains. An iterative weighting mechanism amplifies the influence of consistently reliable evaluators, aggregating peer judgments into consensus-based rankings that reflect collective model agreement. Our experiments demonstrate strong correlations with established benchmarks including MMLU-Pro and GPQA (respectively 78\% and 63\%), validating this peer-driven evaluation paradigm. The multi-judge design significantly outperforms single-judge baselines, confirming that distributed evaluation produces more robust and human-consistent assessments. AutoBench offers a scalable, contamination-resistant alternative to static benchmarks for the continuous evaluation of evolving language models.

Current benchmarks for AI clinician systems, often based on multiple-choice exams or manual rubrics, fail to capture the depth, robustness, and safety required for real-world clinical practice. To address this, we introduce the GAPS framework, a multidimensional paradigm for evaluating Grounding (cognitive depth), Adequacy (answer completeness), Perturbation (robustness), and Safety. Critically, we developed a fully automated, guideline-anchored pipeline to construct a GAPS-aligned benchmark end-to-end, overcoming the scalability and subjectivity limitations of prior work. Our pipeline assembles an evidence neighborhood, creates dual graph and tree representations, and automatically generates questions across G-levels. Rubrics are synthesized by a DeepResearch agent that mimics GRADE-consistent, PICO-driven evidence review in a ReAct loop. Scoring is performed by an ensemble of large language model (LLM) judges. Validation confirmed our automated questions are high-quality and align with clinician judgment. Evaluating state-of-the-art models on the benchmark revealed key failure modes: performance degrades sharply with increased reasoning depth (G-axis), models struggle with answer completeness (A-axis), and they are highly vulnerable to adversarial perturbations (P-axis) as well as certain safety issues (S-axis). This automated, clinically-grounded approach provides a reproducible and scalable method for rigorously evaluating AI clinician systems and guiding their development toward safer, more reliable clinical practice.

The ability to translate diverse patterns of inputs into structured patterns of behavior has been thought to rest on both humans' and machines' ability to learn robust representations of relevant concepts. The rapid advancement of transformer-based large language models (LLMs) has led to a diversity of computational ingredients -- architectures, fine tuning methods, and training datasets among others -- but it remains unclear which of these ingredients are most crucial for building models that develop human-like representations. Further, most current LLM benchmarks are not suited to measuring representational alignment between humans and models, making benchmark scores unreliable for assessing if current LLMs are making progress towards becoming useful cognitive models. We address these limitations by first evaluating a set of over 70 models that widely vary in their computational ingredients on a triplet similarity task, a method well established in the cognitive sciences for measuring human conceptual representations, using concepts from the THINGS database. Comparing human and model representations, we find that models that undergo instruction-finetuning and which have larger dimensionality of attention heads are among the most human aligned, while multimodal pretraining and parameter size have limited bearing on alignment. Correlations between alignment scores and scores on existing benchmarks reveal that while some benchmarks (e.g., MMLU) are better suited than others (e.g., MUSR) for capturing representational alignment, no existing benchmark is capable of fully accounting for the variance of alignment scores, demonstrating their insufficiency in capturing human-AI alignment. Taken together, our findings help highlight the computational ingredients most essential for advancing LLMs towards models of human conceptual representation and address a key benchmarking gap in LLM evaluation.

Scaling test-time computation, or affording a generator large language model (LLM) extra compute during inference, typically employs the help of external non-generative evaluators (i.e., reward models). Concurrently, LLM-judges, models trained to generate evaluations and critiques (explanations) in natural language, are becoming increasingly popular in automatic evaluation. Despite judge empirical successes, their effectiveness as evaluators in test-time scaling settings is largely unknown. In this paper, we introduce the Judge Evaluation for Test-Time Scaling (JETTS) benchmark, which evaluates judge performance in three domains (math reasoning, code generation, and instruction following) under three task settings: response reranking, step-level beam search, and critique-based response refinement. We evaluate 10 different judge models (7B-70B parameters) for 8 different base generator models (6.7B-72B parameters). Our benchmark shows that while judges are competitive with outcome reward models in reranking, they are consistently worse than process reward models in beam search procedures. Furthermore, though unique to LLM-judges, their natural language critiques are currently ineffective in guiding the generator towards better responses.

This paper introduces a novel Deep Researcher architecture designed to generate detailed research reports on complex PhD level topics by addressing the inherent limitations of the Parallel Scaling paradigm. Our system utilizes two key innovations: Sequential Research Plan Refinement via Reflection and a Candidates Crossover algorithm. The sequential refinement process is demonstrated as an efficient method that allows the agent to maintain a centralized Global Research Context, enabling it to look back at current progress, reason about the research plan, and intelligently make changes at runtime. This dynamic adaptation contrasts with parallel approaches, which often suffer from siloed knowledge. The Candidates Crossover algorithm further enhances search efficiency by deploying multiple LLM candidates with varied parameters to explore a larger search space, with their findings synthesized to curate a comprehensive final research response. The process concludes with One Shot Report Generation, ensuring the final document is informed by a unified narrative and high fact density. Powered by the Gemini 2.5 Pro model, our Deep Researcher was evaluated on the DeepResearch Bench, a globally recognized benchmark of 100 doctoral level research tasks. Our architecture achieved an overall score of 46.21, demonstrating superior performance by surpassing leading deep research agents such as Claude Researcher, Nvidia AIQ Research Assistant, Perplexity Research, Kimi Researcher and Grok Deeper Search present on the DeepResearch Bench actively running leaderboard. This performance marginally exceeds our previous work, Static DRA, and reinforces the finding that sequential scaling consistently outperforms the parallel self consistency paradigm.

Recent mixed-policy optimization methods for LLM reasoning that interleave or blend supervised and reinforcement learning signals report improvements over the standard SFT-then-RL pipeline. We show that numerous recently published research papers rely on a faulty baseline caused by two distinct bugs: a CPU-offloaded optimizer bug in DeepSpeed that silently drops intermediate micro-batches during gradient accumulation (affecting multiple downstream frameworks including TRL, OpenRLHF and Llama-Factory), and a loss aggregation bug in OpenRLHF that incorrectly weights per-mini-batch losses. Together they suppress SFT performance, with the optimizer bug accounting for most of the gap and the loss aggregation bug contributing a smaller additional effect. Once corrected, the standard SFT-then-RL pipeline surpasses every published mixed-policy method we evaluate by +3.8 points on math benchmarks with Qwen2.5-Math-7B and by +22.2 points with Llama-3.1-8B. Even a truncated variant with just 50 RL steps outperforms mixed-policy methods on math benchmarks while using fewer FLOPs.

Predictive code completion greatly accelerates how quickly developers work. In spreadsheets, despite being much more common, such auto-completion features are virtually non-existent. To address this gap, we introduce a benchmark for systems that observe a sequence of user actions in a spreadsheet and predict future actions. Two challenges are (1) the absence of edit histories in public spreadsheet corpora and (2) the complex space of spreadsheet actions (spatial, temporal, composite). To address (1), we manually curate 52 sequences of 12K actions that recreate spreadsheets from public corpora, seeded by parametrized heuristics and LLM refinement. To address (2), we propose an online evaluation that expects a prediction after each user action, accepts or rejects that prediction, updates the future actions upon acceptance, and repeats this until the target spreadsheet is obtained. We use multiple baseline predictors (including zero-shot LLMs, fine-tuned SLMs, and classical models) and analyze different properties that our benchmark teaches us, including but not limited to: properties of saved actions and false positives, efficiency, effect of user profiles, effect of triggers, and effect of context.

Benchmark datasets for table structure recognition (TSR) must be carefully processed to ensure they are annotated consistently. However, even if a dataset's annotations are self-consistent, there may be significant inconsistency across datasets, which can harm the performance of models trained and evaluated on them. In this work, we show that aligning these benchmarksx2014removing both errors and inconsistency between themx2014improves model performance significantly. We demonstrate this through a data-centric approach where we adopt a single model architecture, the Table Transformer (TATR), that we hold fixed throughout. Baseline exact match accuracy for TATR evaluated on the ICDAR-2013 benchmark is 65% when trained on PubTables-1M, 42% when trained on FinTabNet, and 69% combined. After reducing annotation mistakes and inter-dataset inconsistency, performance of TATR evaluated on ICDAR-2013 increases substantially to 75% when trained on PubTables-1M, 65% when trained on FinTabNet, and 81% combined. We show through ablations over the modification steps that canonicalization of the table annotations has a significantly positive effect on performance, while other choices balance necessary trade-offs that arise when deciding a benchmark dataset's final composition. Overall we believe our work has significant implications for benchmark design for TSR and potentially other tasks as well. All dataset processing and training code will be released.

The rapid release of both language models and benchmarks makes it increasingly costly to evaluate every model on every dataset. In practice, models are often evaluated on different samples, making scores difficult to compare across studies. To address this, we propose a framework based on multidimensional Item Response Theory (IRT) that uses anchor items to calibrate new benchmarks to the evaluation suite while holding previously calibrated item parameters fixed. Our approach supports a realistic evaluation setting in which datasets are introduced over time and models are evaluated only on the datasets available at the time of evaluation, while a fixed anchor set for each dataset is used so that results from different evaluation periods can be compared directly. In large-scale experiments on more than 400 models, our framework predicts full-evaluation performance within 2-3 percentage points using only 100 anchor questions per dataset, with Spearman ρgeq 0.9 for ranking preservation, showing that it is possible to extend benchmark suites over time while preserving score comparability, at a constant evaluation cost per new dataset. Code available at https://github.com/eliyahabba/growing-pains

There is a consensus that instruction fine-tuning of LLMs requires high-quality data, but what are they? LIMA (NeurIPS 2023) and AlpaGasus (ICLR 2024) are state-of-the-art methods for selecting such high-quality examples, either via manual curation or using GPT-3.5-Turbo as a quality scorer. We show that the extremely simple baseline of selecting the 1,000 instructions with longest responses from standard datasets can consistently outperform these sophisticated methods according to GPT-4 and PaLM-2 as judges, while remaining competitive on the OpenLLM benchmarks that test factual knowledge. We demonstrate this for several state-of-the-art LLMs (Llama-2-7B, Llama-2-13B, and Mistral-7B) and datasets (Alpaca-52k and Evol-Instruct-70k). In addition, a lightweight refinement of such long instructions can further improve the abilities of the fine-tuned LLMs, and allows us to obtain the 2nd highest-ranked Llama-2-7B-based model on AlpacaEval 2.0 while training on only 1,000 examples and no extra preference data. We also conduct a thorough analysis of our models to ensure that their enhanced performance is not simply due to GPT-4's preference for longer responses, thus ruling out any artificial improvement. In conclusion, our findings suggest that fine-tuning on the longest instructions should be the default baseline for any research on instruction fine-tuning.

Recent machine learning papers often report 1-2 percentage point improvements from a single run on a benchmark. These gains are highly sensitive to random seeds, data ordering, and implementation details, yet are rarely accompanied by uncertainty estimates or significance tests. It is therefore unclear when a reported +1-2% reflects a real algorithmic advance versus noise. We revisit this problem under realistic compute budgets, where only a few runs are affordable. We propose a simple, PC-friendly evaluation protocol based on paired multi-seed runs, bias-corrected and accelerated (BCa) bootstrap confidence intervals, and a sign-flip permutation test on per-seed deltas. The protocol is intentionally conservative and is meant as a guardrail against over-claiming. We instantiate it on CIFAR-10, CIFAR-10N, and AG News using synthetic no-improvement, small-gain, and medium-gain scenarios. Single runs and unpaired t-tests often suggest significant gains for 0.6-2.0 point improvements, especially on text. With only three seeds, our paired protocol never declares significance in these settings. We argue that such conservative evaluation is a safer default for small gains under tight budgets.

Evaluation benchmarks are the cornerstone of measuring capabilities of large language models (LLMs), as well as driving progress in said capabilities. Originally designed to make claims about capabilities (or lack thereof) in fully pretrained models, evaluation benchmarks are now also extensively used to decide between various training choices. Despite this widespread usage, we rarely quantify the variance in our evaluation benchmarks, which dictates whether differences in performance are meaningful. Here, we define and measure a range of metrics geared towards measuring variance in evaluation benchmarks, including seed variance across initialisations, and monotonicity during training. By studying a large number of models -- both openly available and pretrained from scratch -- we provide empirical estimates for a variety of variance metrics, with considerations and recommendations for practitioners. We also evaluate the utility and tradeoffs of continuous versus discrete performance measures and explore options for better understanding and reducing this variance. We find that simple changes, such as framing choice tasks (like MMLU) as completion tasks, can often reduce variance for smaller scale (sim7B) models, while more involved methods inspired from human testing literature (such as item analysis and item response theory) struggle to meaningfully reduce variance. Overall, our work provides insights into variance in evaluation benchmarks, suggests LM-specific techniques to reduce variance, and more generally encourages practitioners to carefully factor in variance when comparing models.

The correct use of model evaluation, model selection, and algorithm selection techniques is vital in academic machine learning research as well as in many industrial settings. This article reviews different techniques that can be used for each of these three subtasks and discusses the main advantages and disadvantages of each technique with references to theoretical and empirical studies. Further, recommendations are given to encourage best yet feasible practices in research and applications of machine learning. Common methods such as the holdout method for model evaluation and selection are covered, which are not recommended when working with small datasets. Different flavors of the bootstrap technique are introduced for estimating the uncertainty of performance estimates, as an alternative to confidence intervals via normal approximation if bootstrapping is computationally feasible. Common cross-validation techniques such as leave-one-out cross-validation and k-fold cross-validation are reviewed, the bias-variance trade-off for choosing k is discussed, and practical tips for the optimal choice of k are given based on empirical evidence. Different statistical tests for algorithm comparisons are presented, and strategies for dealing with multiple comparisons such as omnibus tests and multiple-comparison corrections are discussed. Finally, alternative methods for algorithm selection, such as the combined F-test 5x2 cross-validation and nested cross-validation, are recommended for comparing machine learning algorithms when datasets are small.

The prohibitive cost of evaluating large language models (LLMs) on comprehensive benchmarks necessitates the creation of small yet representative data subsets (i.e., tiny benchmarks) that enable efficient assessment while retaining predictive fidelity. Current methods for this task operate under a model-centric paradigm, selecting benchmarking items based on the collective performance of existing models. Such approaches are limited by large upfront costs, an inability to immediately handle new benchmarks (`cold-start'), and the fragile assumption that future models will share the failure patterns of their predecessors. In this work, we challenge this paradigm and propose a item-centric approach to benchmark subset selection, arguing that selection should be based on the intrinsic properties of the task items themselves, rather than on model-specific failure patterns. We instantiate this item-centric efficient benchmarking approach via a novel method, Scales++, where data selection is based on the cognitive demands of the benchmark samples. Empirically, we show Scales++ reduces the upfront selection cost by over 18x while achieving competitive predictive fidelity. On the Open LLM Leaderboard, using just a 0.5\% data subset, we predict full benchmark scores with a 2.9% mean absolute error. We demonstrate that this item-centric approach enables more efficient model evaluation without significant fidelity degradation, while also providing better cold-start performance and more interpretable benchmarking.

Recent observations have underscored a disparity between the inflated benchmark scores and the actual performance of LLMs, raising concerns about potential contamination of evaluation benchmarks. This issue is especially critical for closed-source models and certain open-source models where training data transparency is lacking. In this paper we study data contamination by proposing two methods tailored for both open-source and proprietary LLMs. We first introduce a retrieval-based system to explore potential overlaps between evaluation benchmarks and pretraining corpora. We further present a novel investigation protocol named Testset Slot Guessing (TS-Guessing), applicable to both open and proprietary models. This approach entails masking a wrong answer in a multiple-choice question and prompting the model to fill in the gap. Additionally, it involves obscuring an unlikely word in an evaluation example and asking the model to produce it. We find that certain commercial LLMs could surprisingly guess the missing option in various test sets. Specifically, in the TruthfulQA benchmark, we find that LLMs exhibit notable performance improvement when provided with additional metadata in the benchmark. Further, in the MMLU benchmark, ChatGPT and GPT-4 demonstrated an exact match rate of 52\% and 57\%, respectively, in guessing the missing options in benchmark test data. We hope these results underscore the need for more robust evaluation methodologies and benchmarks in the field.

Recent advances in self-supervised models for natural language, vision, and protein sequences have inspired the development of large genomic DNA language models (DNALMs). These models aim to learn generalizable representations of diverse DNA elements, potentially enabling various genomic prediction, interpretation and design tasks. Despite their potential, existing benchmarks do not adequately assess the capabilities of DNALMs on key downstream applications involving an important class of non-coding DNA elements critical for regulating gene activity. In this study, we introduce DART-Eval, a suite of representative benchmarks specifically focused on regulatory DNA to evaluate model performance across zero-shot, probed, and fine-tuned scenarios against contemporary ab initio models as baselines. Our benchmarks target biologically meaningful downstream tasks such as functional sequence feature discovery, predicting cell-type specific regulatory activity, and counterfactual prediction of the impacts of genetic variants. We find that current DNALMs exhibit inconsistent performance and do not offer compelling gains over alternative baseline models for most tasks, while requiring significantly more computational resources. We discuss potentially promising modeling, data curation, and evaluation strategies for the next generation of DNALMs. Our code is available at https://github.com/kundajelab/DART-Eval.

Fine-tuning of pre-trained transformer models has become the standard approach for solving common NLP tasks. Most of the existing approaches rely on a randomly initialized classifier on top of such networks. We argue that this fine-tuning procedure is sub-optimal as the pre-trained model has no prior on the specific classifier labels, while it might have already learned an intrinsic textual representation of the task. In this paper, we introduce a new scoring method that casts a plausibility ranking task in a full-text format and leverages the masked language modeling head tuned during the pre-training phase. We study commonsense reasoning tasks where the model must rank a set of hypotheses given a premise, focusing on the COPA, Swag, HellaSwag and CommonsenseQA datasets. By exploiting our scoring method without fine-tuning, we are able to produce strong baselines (e.g. 80% test accuracy on COPA) that are comparable to supervised approaches. Moreover, when fine-tuning directly on the proposed scoring function, we show that our method provides a much more stable training phase across random restarts (e.g times 10 standard deviation reduction on COPA test accuracy) and requires less annotated data than the standard classifier approach to reach equivalent performances.

Large language models have made significant progress in complex but easy-to-verify problems, yet they still struggle with discovering the unknown. In this paper, we present AlphaResearch, an autonomous research agent designed to discover new algorithms on open-ended problems. To synergize the feasibility and innovation of the discovery process, we construct a novel dual research environment by combining the execution-based verify and simulated real-world peer review environment. AlphaResearch discovers new algorithm by iteratively running the following steps: (1) propose new ideas (2) verify the ideas in the dual research environment (3) optimize the research proposals for better performance. To promote a transparent evaluation process, we construct AlphaResearchComp, a new evaluation benchmark that includes an eight open-ended algorithmic problems competition, with each problem carefully curated and verified through executable pipelines, objective metrics, and reproducibility checks. AlphaResearch gets a 2/8 win rate in head-to-head comparison with human researchers, demonstrate the possibility of accelerating algorithm discovery with LLMs. Notably, the algorithm discovered by AlphaResearch on the ``packing circles'' problem achieves the best-of-known performance, surpassing the results of human researchers and strong baselines from recent work (e.g., AlphaEvolve). Additionally, we conduct a comprehensive analysis of the remaining challenges of the 6/8 failure cases, providing valuable insights for future research.

In recent years, while natural language processing and multimodal learning have seen rapid advancements, the field of de novo protein design has also experienced significant growth. However, most current methods rely on proprietary datasets and evaluation rubrics, making fair comparisons between different approaches challenging. Moreover, these methods often employ evaluation metrics that capture only a subset of the desired properties of designed proteins, lacking a comprehensive assessment framework. To address these, we introduce PDFBench, the first comprehensive benchmark for evaluating de novo protein design from function. PDFBench supports two tasks: description-guided design and keyword-guided design. To ensure fair and multifaceted evaluation, we compile 22 metrics covering sequence plausibility, structural fidelity, and language-protein alignment, along with measures of novelty and diversity. We evaluate five state-of-the-art baselines, revealing their respective strengths and weaknesses across tasks. Finally, we analyze inter-metric correlations, exploring the relationships between four categories of metrics, and offering guidelines for metric selection. PDFBench establishes a unified framework to drive future advances in function-driven de novo protein design.

With the rapid advancement of large language models (LLMs), there is a pressing need for a comprehensive evaluation suite to assess their capabilities and limitations. Existing LLM leaderboards often reference scores reported in other papers without consistent settings and prompts, which may inadvertently encourage cherry-picking favored settings and prompts for better results. In this work, we introduce GPT-Fathom, an open-source and reproducible LLM evaluation suite built on top of OpenAI Evals. We systematically evaluate 10+ leading LLMs as well as OpenAI's legacy models on 20+ curated benchmarks across 7 capability categories, all under aligned settings. Our retrospective study on OpenAI's earlier models offers valuable insights into the evolutionary path from GPT-3 to GPT-4. Currently, the community is eager to know how GPT-3 progressively improves to GPT-4, including technical details like whether adding code data improves LLM's reasoning capability, which aspects of LLM capability can be improved by SFT and RLHF, how much is the alignment tax, etc. Our analysis sheds light on many of these questions, aiming to improve the transparency of advanced LLMs.

Distinct-n scoreLi2016 is a widely used automatic metric for evaluating diversity in language generation tasks. However, we observed that the original approach for calculating distinct scores has evident biases that tend to assign higher penalties to longer sequences. We refine the calculation of distinct scores by scaling the number of distinct tokens based on their expectations. We provide both empirical and theoretical evidence to show that our method effectively removes the biases existing in the original distinct score. Our experiments show that our proposed metric, Expectation-Adjusted Distinct (EAD), correlates better with human judgment in evaluating response diversity. To foster future research, we provide an example implementation at https://github.com/lsy641/Expectation-Adjusted-Distinct.

Static benchmarks measure a model frozen at training time. Real systems face distribution shift: new categories, paraphrased queries, drift: and must recover online via user corrections. No existing benchmark measures recovery speed under correction streams. We introduce OCRR (Online Correction Recovery Rate): a benchmark that streams a corpus through a classification system, applies oracle or stochastic corrections to wrong predictions, and reports two curves: novel-class accuracy and original-distribution accuracy versus correction count. We evaluate the substrate alongside nine baseline algorithms from five families plus seven bounded-storage variants of the substrate for the Pareto sweep, including standard online-learning baselines (river), continual-learning methods (EWC, A-GEM, LwF), retrieval/parametric hybrids (kNN-LM), parameter-efficient fine-tuning of a 1.5 B-parameter encoder (LoRA on DeBERTa-v3-large), and a hash-chained append-only substrate (Substrate). On Banking77 and CLINC150, under oracle and sparse correction policies, the substrate is the only system that simultaneously recovers novel-class accuracy (88.7 +/- 2.9 %) and retains original-distribution accuracy (95.4 +/- 0.8 %) beating the next-best published continual-learning baseline by 32.6 percentage points at equal memory budget, and beating LoRA-on-DeBERTa-v3-large by 84.6 percentage points on retention. We further find that classification accuracy remains stable at 99 % even as approximate-nearest-neighbour recall@5 degrades from 0.69 to 0.23 across 10 k to 10 M corpus scales, suggesting the substrate's margin-band majority vote is robust to retrieval imperfection in a way that pure top-k recall metrics do not predict. Code and data are available at https://github.com/adriangrassi/ocrr-benchmark.

Large Language Models (LLMs) are increasingly used for clinical decision support, where hallucinations and unsafe suggestions may pose direct risks to patient safety. These risks are particularly challenging as they often manifest as subtle clinical errors that evade detection by generic metrics, while expert-authored fine-grained rubrics remain costly to construct and difficult to scale. In this paper, we propose a retrieval-augmented multi-agent framework designed to automate the generation of instance-specific evaluation rubrics. Our approach grounds evaluation in authoritative medical evidence by decomposing retrieved content into atomic facts and synthesizing them with user interaction constraints to form verifiable, fine-grained evaluation criteria. Evaluated on HealthBench, our framework achieves a Clinical Intent Alignment (CIA) score of 60.12%, a statistically significant improvement over the GPT-4o baseline (55.16%). In discriminative tests, our rubrics yield a mean score delta (μ_Δ = 8.658) and an AUROC of 0.977, nearly doubling the quality separation achieved by GPT-4o baseline (4.972). Beyond evaluation, our rubrics effectively guide response refinement, improving quality by 9.2% (from 59.0% to 68.2%). This provides a scalable and transparent foundation for both evaluating and improving medical LLMs. The code is available at https://anonymous.4open.science/r/Automated-Rubric-Generation-AF3C/.

We present our winning system for Task~B (generation with reference passages) in SemEval-2026 Task~8: MTRAGEval. Our method is a heterogeneous ensemble of seven LLMs with two prompting variants, where a GPT-4o-mini judge selects the best candidate per instance. We ranked 1st out of 26 teams, achieving a conditioned harmonic mean of 0.7827 and outperforming the strongest baseline (gpt-oss-120b, 0.6390). Ablations show that diversity in model families, scales, and prompting strategies is essential, with the ensemble consistently beating any single model. We also introduce Meno-Lite-0.1, a 7B domain-adapted model with a strong cost--performance trade-off, and analyse MTRAGEval, highlighting annotation limitations and directions for improvement. Our code is publicly available: https://github.com/RaguTeam/ragu_mtrag_semeval

Because large language models are expensive to pretrain on different datasets, using smaller-scale experiments to decide on data is crucial for reducing costs. Which benchmarks and methods of making decisions from observed performance at small scale most accurately predict the datasets that yield the best large models? To empower open exploration of this question, we release models, data, and evaluations in DataDecide -- the most extensive open suite of models over differences in data and scale. We conduct controlled pretraining experiments across 25 corpora with differing sources, deduplication, and filtering up to 100B tokens, model sizes up to 1B parameters, and 3 random seeds. We find that the ranking of models at a single, small size (e.g., 150M parameters) is a strong baseline for predicting best models at our larger target scale (1B) (~80% of com parisons correct). No scaling law methods among 8 baselines exceed the compute-decision frontier of single-scale predictions, but DataDecide can measure improvement in future scaling laws. We also identify that using continuous likelihood metrics as proxies in small experiments makes benchmarks including MMLU, ARC, HellaSwag, MBPP, and HumanEval >80% predictable at the target 1B scale with just 0.01% of the compute.

Large language models (LLMs) have achieved impressive results on multi-step mathematical reasoning, yet at the cost of high computational overhead. This challenge is particularly acute for test-time scaling methods such as parallel decoding, which increase answer diversity but scale poorly in efficiency. To address this efficiency-accuracy trade-off, we propose SSR (Speculative Parallel Scaling Reasoning), a training-free framework that leverages a key insight: by introducing speculative decoding at the step level, we can accelerate reasoning without sacrificing correctness. SSR integrates two components: a Selective Parallel Module (SPM) that identifies a small set of promising reasoning strategies via model-internal scoring, and Step-level Speculative Decoding (SSD), which enables efficient draft-target collaboration for fine-grained reasoning acceleration. Experiments on three mathematical benchmarks-AIME 2024, MATH-500, and LiveMathBench - demonstrate that SSR achieves strong gains over baselines. For instance, on LiveMathBench, SSR improves pass@1 accuracy by 13.84% while reducing computation to 80.5% of the baseline FLOPs. On MATH-500, SSR reduces compute to only 30% with no loss in accuracy.

Retrieval-augmented generation (RAG) has emerged as a leading approach to reducing hallucinations in large language models (LLMs). Current RAG evaluation benchmarks primarily focus on what we call local RAG: retrieving relevant chunks from a small subset of documents to answer queries that require only localized understanding within specific text chunks. However, many real-world applications require a fundamentally different capability -- global RAG -- which involves aggregating and analyzing information across entire document collections to derive corpus-level insights (for example, "What are the top 10 most cited papers in 2023?"). In this paper, we introduce GlobalQA -- the first benchmark specifically designed to evaluate global RAG capabilities, covering four core task types: counting, extremum queries, sorting, and top-k extraction. Through systematic evaluation across different models and baselines, we find that existing RAG methods perform poorly on global tasks, with the strongest baseline achieving only 1.51 F1 score. To address these challenges, we propose GlobalRAG, a multi-tool collaborative framework that preserves structural coherence through chunk-level retrieval, incorporates LLM-driven intelligent filters to eliminate noisy documents, and integrates aggregation modules for precise symbolic computation. On the Qwen2.5-14B model, GlobalRAG achieves 6.63 F1 compared to the strongest baseline's 1.51 F1, validating the effectiveness of our method.

Automatic evaluation metrics are essential for building multilingual translation systems. The common practice of evaluating these systems is averaging metric scores across languages, yet this is suspicious since metrics may suffer from cross-lingual scoring bias, where translations of equal quality receive different scores across languages. This problem has not been systematically studied because no benchmark exists that provides parallel-quality instances across languages, and expert annotation is not realistic. In this work, we propose XQ-MEval, a semi-automatically built dataset covering nine translation directions, to benchmark translation metrics. Specifically, we inject MQM-defined errors into gold translations automatically, filter them by native speakers for reliability, and merge errors to generate pseudo translations with controllable quality. These pseudo translations are then paired with corresponding sources and references to form triplets used in assessing the qualities of translation metrics. Using XQ-MEval, our experiments on nine representative metrics reveal the inconsistency between averaging and human judgment and provide the first empirical evidence of cross-lingual scoring bias. Finally, we propose a normalization strategy derived from XQ-MEval that aligns score distributions across languages, improving the fairness and reliability of multilingual metric evaluation.

As LLM benchmarks saturate, the evaluation community has pursued two strategies to increase difficulty: escalating knowledge demands (GPQA, HLE) or removing knowledge entirely in favor of abstract reasoning (ARC-AGI). The first conflates memorization with capability; the second divorces reasoning from the practical contexts in which it matters. We take a different approach. The Grounded Integration Measure (GIM) is a benchmark of 820 original problems (615 public, 205 private) where difficulty comes from integration; individual problems require coordinating multiple cognitive operations (constraint satisfaction, state tracking, epistemic vigilance, audience calibration) over broadly accessible knowledge, so that reasoning stays grounded in realistic tasks without being gated on specialized expertise. Each problem is an original expert-authored composition, majority with rubric-decomposed scoring (median 6 independently judged criteria). A balanced public--private split provides built-in contamination diagnostic. We calibrate a continuous response 2-parameter logistic (2PL) IRT model over >200k prompt-response pairs across 28 models, producing robust ability estimates that correctly order test-configurations even when raw accuracy is distorted by errors or missing data, addressing a common challenge in benchmark reporting. Using this framework, we present a comprehensive leaderboard spanning 22 models and 47 test-configurations (unique model, thinking-level pairs), and conduct what is to our knowledge the most extensive published study of how test-time compute trades off against model capability on a fixed benchmark: 11 models swept across 35 test-configurations. We observe that within-family configuration choices, such as thinking budget and quantization, matter as much as model selection. We release the evaluation framework, calibrated IRT parameters, and all public problems.

The evaluation of Large Language Models (LLMs) increasingly relies on other LLMs acting as judges. However, current evaluation paradigms typically yield a single score or ranking, answering which model is better but not why. While essential for benchmarking, these top-level scores obscure the specific, actionable reasons behind a model's performance. To bridge this gap, we introduce CLEAR, an interactive, open-source package for LLM-based error analysis. CLEAR first generates per-instance textual feedback, then it creates a set of system-level error issues, and quantifies the prevalence of each identified issue. Our package also provides users with an interactive dashboard that allows for a comprehensive error analysis through aggregate visualizations, applies interactive filters to isolate specific issues or score ranges, and drills down to the individual instances that exemplify a particular behavioral pattern. We demonstrate CLEAR analysis for RAG and Math benchmarks, and showcase its utility through a user case study.

Multi-turn-to-single-turn (M2S) compresses iterative red-teaming into one structured prompt, but prior work relied on a handful of manually written templates. We present X-Teaming Evolutionary M2S, an automated framework that discovers and optimizes M2S templates through language-model-guided evolution. The system pairs smart sampling from 12 sources with an LLM-as-judge inspired by StrongREJECT and records fully auditable logs. Maintaining selection pressure by setting the success threshold to theta = 0.70, we obtain five evolutionary generations, two new template families, and 44.8% overall success (103/230) on GPT-4.1. A balanced cross-model panel of 2,500 trials (judge fixed) shows that structural gains transfer but vary by target; two models score zero at the same threshold. We also find a positive coupling between prompt length and score, motivating length-aware judging. Our results demonstrate that structure-level search is a reproducible route to stronger single-turn probes and underscore the importance of threshold calibration and cross-model evaluation. Code, configurations, and artifacts are available at https://github.com/hyunjun1121/M2S-x-teaming.

Large Language Models (LLMs) rely on Human Preference Alignment (HPA) to ensure the generation of safe content. Due to the heavy cost associated with fine-tuning, fine-tuning-free methods have emerged, typically modifying LLM decoding with external auxiliary methods. However, these methods do not essentially enhance the LLM itself. In this paper, we rethink the derivation procedures of DPO, based on which we conversely build an instant scorer using the states of the LLM before and after In-context Learning (ICL). Accordingly, we propose a novel approach called In-Context Direct Preference Optimization (ICDPO). It enables LLMs to borrow the HPA capabilities from superior LLMs with ICL, generating well-aligned responses as estimated by the aforementioned instant scorer, thereby enhancing the final performance. ICDPO can be further enhanced with a two-stage retriever and an upgraded scorer, both offering benefits. Extensive experiments show its effectiveness, particularly in outperforming two fine-tuning-free baselines, and it exhibits competitiveness with SFT + LoRA. We also conduct detailed analyses to offer comprehensive insights into ICDPO.

Hampering the interpretation of benchmark scores, evaluation data contamination has become a growing concern in the evaluation of LLMs, and an active area of research studies its effects. While evaluation data contamination is easily understood intuitively, it is surprisingly difficult to define precisely which samples should be considered contaminated and, consequently, how it impacts benchmark scores. We propose that these questions should be addressed together and that contamination metrics can be assessed based on whether models benefit from the examples they mark contaminated. We propose a novel analysis method called ConTAM, and show with a large scale survey of existing and novel n-gram based contamination metrics across 13 benchmarks and 7 models from 2 different families that ConTAM can be used to better understand evaluation data contamination and its effects. We find that contamination may have a much larger effect than reported in recent LLM releases and benefits models differently at different scales. We also find that considering only the longest contaminated substring provides a better signal than considering a union of all contaminated substrings, and that doing model and benchmark specific threshold analysis greatly increases the specificity of the results. Lastly, we investigate the impact of hyperparameter choices, finding that, among other things, both using larger values of n and disregarding matches that are infrequent in the pre-training data lead to many false negatives. With ConTAM, we provide a method to empirically ground evaluation data contamination metrics in downstream effects. With our exploration, we shed light on how evaluation data contamination can impact LLMs and provide insight into the considerations important when doing contamination analysis. We end our paper by discussing these in more detail and providing concrete suggestions for future work.

Aligning large visual generative models with human feedback is often performed through pairwise preference optimization. While such approaches are conceptually simple, they fundamentally rely on annotated pairs, limiting scalability in settings where feedback is collected as independent scalar ratings. In this work, we revisit the KL-regularized alignment objective and show that the optimal policy implicitly compares each sample's reward to an instance-specific baseline that is generally intractable. We propose a threshold-guided alignment framework that replaces this oracle baseline with a data-driven global threshold estimated from empirical score statistics. This formulation turns alignment into a binary decision task on unpaired data, enabling effective optimization directly from scalar feedback. We also incorporate a confidence weighting term to emphasize samples whose scores deviate strongly from the threshold, improving sample efficiency. Experiments across both diffusion and masked generative paradigms, spanning three test sets and five reward models, show that our method consistently improves preference alignment over previous methods. These results position our threshold-guided framework as a simple yet principled alternative for aligning visual generative models without paired comparisons.

Recent advances in Large Language Models (LLMs) and Reinforcement Learning (RL) have led to strong performance in open-domain question answering (QA). However, existing models still struggle with questions that admit multiple valid answers. Standard QA benchmarks, which typically assume a single gold answer, overlook this reality and thus produce inappropriate training signals. Existing attempts to handle ambiguity often rely on costly manual annotation, which is difficult to scale to multi-hop datasets such as HotpotQA and MuSiQue. In this paper, we present A^2Search, an annotation-free, end-to-end training framework to recognize and handle ambiguity. At its core is an automated pipeline that detects ambiguous questions and gathers alternative answers via trajectory sampling and evidence verification. The model is then optimized with RL using a carefully designed AnsF1 reward, which naturally accommodates multiple answers. Experiments on eight open-domain QA benchmarks demonstrate that A^2Search achieves new state-of-the-art performance. With only a single rollout, A^2Search-7B yields an average AnsF1@1 score of 48.4% across four multi-hop benchmarks, outperforming all strong baselines, including the substantially larger ReSearch-32B (46.2%). Extensive analyses further show that A^2Search resolves ambiguity and generalizes across benchmarks, highlighting that embracing ambiguity is essential for building more reliable QA systems. Our code, data, and model weights can be found at https://github.com/zfj1998/A2Search

Many benchmarks for automated causal inference evaluate a system's performance based on a single numerical output, such as an Average Treatment Effect (ATE). This approach conflates two distinct steps in causal analysis: identification-formulating a valid research design under stated assumptions-and estimation-implementing that design numerically on finite data. We introduce CausalReasoningBenchmark, a benchmark of 173 queries across 138 real-world datasets, curated from 85 peer-reviewed research papers and four widely-used causal-inference textbooks. For each query a system must produce (i) a structured identification specification that names the strategy, the treatment, outcome, and control variables, and all design-specific elements, and (ii) a point estimate with a standard error. By scoring these two components separately, our benchmark enables granular diagnosis: it distinguishes failures in causal reasoning from errors in numerical execution. Baseline results with a state-of-the-art LLM show that, while the model correctly identifies the high-level strategy in 84 % of cases, full identification-specification correctness drops to only 30 %, revealing that the bottleneck lies in the nuanced details of research design rather than in computation. CausalReasoningBenchmark is publicly available on Hugging Face and is designed to foster the development of more robust automated causal-inference systems.

Multimodal/vision language models (VLMs) are increasingly being deployed in healthcare settings worldwide, necessitating robust benchmarks to ensure their safety, efficacy, and fairness. Multiple-choice question and answer (QA) datasets derived from national medical examinations have long served as valuable evaluation tools, but existing datasets are largely text-only and available in a limited subset of languages and countries. To address these challenges, we present WorldMedQA-V, an updated multilingual, multimodal benchmarking dataset designed to evaluate VLMs in healthcare. WorldMedQA-V includes 568 labeled multiple-choice QAs paired with 568 medical images from four countries (Brazil, Israel, Japan, and Spain), covering original languages and validated English translations by native clinicians, respectively. Baseline performance for common open- and closed-source models are provided in the local language and English translations, and with and without images provided to the model. The WorldMedQA-V benchmark aims to better match AI systems to the diverse healthcare environments in which they are deployed, fostering more equitable, effective, and representative applications.

The Languini Kitchen serves as both a research collective and codebase designed to empower researchers with limited computational resources to contribute meaningfully to the field of language modelling. We introduce an experimental protocol that enables model comparisons based on equivalent compute, measured in accelerator hours. The number of tokens on which a model is trained is defined by the model's throughput and the chosen compute class. Notably, this approach avoids constraints on critical hyperparameters which affect total parameters or floating-point operations. For evaluation, we pre-process an existing large, diverse, and high-quality dataset of books that surpasses existing academic benchmarks in quality, diversity, and document length. On it, we compare methods based on their empirical scaling trends which are estimated through experiments at various levels of compute. This work also provides two baseline models: a feed-forward model derived from the GPT-2 architecture and a recurrent model in the form of a novel LSTM with ten-fold throughput. While the GPT baseline achieves better perplexity throughout all our levels of compute, our LSTM baseline exhibits a predictable and more favourable scaling law. This is due to the improved throughput and the need for fewer training tokens to achieve the same decrease in test perplexity. Extrapolating the scaling laws leads of both models results in an intersection at roughly 50,000 accelerator hours. We hope this work can serve as the foundation for meaningful and reproducible language modelling research.

Emerging personal AI agents are moving toward persistent, multi-source memory. This creates an evaluation problem: systems must decide how to use conflicting or incomplete evidence; they cannot just retrieve facts from one clean history. Existing benchmarks rarely show whether an error came from the evidence given to a method or from the method's conflict-resolution step. We study this as selective QA over conflicting multi-source personal memory: systems answer based on conflicting, sometimes incomplete sources, or abstain when evidence is insufficient. We develop a benchmark containing 18 question templates across 8 reasoning types, 480 personas, 4 random seeds, and 34,560 instances, with controlled source distortions and deterministic ground truth. We evaluate the performance of baselines without access to any source, access to a single source, structured fusion methods, and frontier LLMs. The best trained fusion resolver reaches 80.3% accuracy, while the strongest prompt-only LLM baseline reaches 70.0%. With abstention, the same resolver reaches 85.3% selective accuracy at 78.3% coverage and the best LLM reaches 71.0% selective accuracy at 95.4% coverage. Different models have different strengths across reasoning types. We release the data, code, cached model outputs, and data-generating process for reuse.

Evaluating whether multimodal large language models truly understand long-form scientific papers remains challenging: answer-only metrics and synthetic "Needle-In-A-Haystack" tests often reward answer matching without requiring a causal, evidence-linked reasoning trace in the document. We propose the "Fish-in-the-Ocean" (FITO) paradigm, which requires models to construct explicit cross-modal evidence chains within native scientific documents. To operationalize FITO, we build SIN-Data, a scientific interleaved corpus that preserves the native interleaving of text and figures. On top of it, we construct SIN-Bench with four progressive tasks covering evidence discovery (SIN-Find), hypothesis verification (SIN-Verify), grounded QA (SIN-QA), and evidence-anchored synthesis (SIN-Summary). We further introduce "No Evidence, No Score", scoring predictions when grounded to verifiable anchors and diagnosing evidence quality via matching, relevance, and logic. Experiments on eight MLLMs show that grounding is the primary bottleneck: Gemini-3-pro achieves the best average overall score (0.573), while GPT-5 attains the highest SIN-QA answer accuracy (0.767) but underperforms on evidence-aligned overall scores, exposing a gap between correctness and traceable support.

Benchmarking competitions are central to the development of artificial intelligence (AI) in medical imaging, defining performance standards and shaping methodological progress. However, it remains unclear whether these benchmarks provide data that are sufficiently representative, accessible, and reusable to support clinically meaningful AI. In this work, we assess fairness along two complementary dimensions: (1) whether challenge datasets are representative of real-world clinical diversity, and (2) whether they are accessible and legally reusable in line with the FAIR principles. To address this question, we conducted a large-scale systematic study of 241 biomedical image analysis challenges comprising 458 tasks across 19 imaging modalities. Our findings show substantial biases in dataset composition, including geographic location, modality-, and problem type-related biases, indicating that current benchmarks do not adequately reflect real-world clinical diversity. Despite their widespread influence, challenge datasets were frequently constrained by restrictive or ambiguous access conditions, inconsistent or non-compliant licensing practices, and incomplete documentation, limiting reproducibility and long-term reuse. Together, these shortcomings expose foundational fairness limitations in our benchmarking ecosystem and highlight a disconnect between leaderboard success and clinical relevance.

Offering a promising solution to the scalability challenges associated with human evaluation, the LLM-as-a-judge paradigm is rapidly gaining traction as an approach to evaluating large language models (LLMs). However, there are still many open questions about the strengths and weaknesses of this paradigm, and what potential biases it may hold. In this paper, we present a comprehensive study of the performance of various LLMs acting as judges. We leverage TriviaQA as a benchmark for assessing objective knowledge reasoning of LLMs and evaluate them alongside human annotations which we found to have a high inter-annotator agreement. Our study includes 9 judge models and 9 exam taker models -- both base and instruction-tuned. We assess the judge model's alignment across different model sizes, families, and judge prompts. Among other results, our research rediscovers the importance of using Cohen's kappa as a metric of alignment as opposed to simple percent agreement, showing that judges with high percent agreement can still assign vastly different scores. We find that both Llama-3 70B and GPT-4 Turbo have an excellent alignment with humans, but in terms of ranking exam taker models, they are outperformed by both JudgeLM-7B and the lexical judge Contains, which have up to 34 points lower human alignment. Through error analysis and various other studies, including the effects of instruction length and leniency bias, we hope to provide valuable lessons for using LLMs as judges in the future.

Software engineers mainly write code by editing existing programs. In contrast, large language models (LLMs) autoregressively synthesize programs in a single pass. One explanation for this is the scarcity of open-sourced edit data. While high-quality instruction data for code synthesis is already scarce, high-quality edit data is even scarcer. To fill this gap, we develop a synthetic data generation algorithm called LintSeq. This algorithm refactors existing code into a sequence of code edits by using a linter to procedurally sample across the error-free insertions that can be used to sequentially write programs. It outputs edit sequences as text strings consisting of consecutive program diffs. To test LintSeq, we use it to refactor a dataset of instruction + program pairs into instruction + program-diff-sequence tuples. Then, we instruction finetune a series of smaller LLMs ranging from 2.6B to 14B parameters on both the re-factored and original versions of this dataset, comparing zero-shot performance on code synthesis benchmarks. We show that during repeated sampling, edit sequence finetuned models produce more diverse programs than baselines. This results in better inference-time scaling for benchmark coverage as a function of samples, i.e. the fraction of problems "pass@k" solved by any attempt given "k" tries. For example, on HumanEval pass@50, small LLMs finetuned on synthetic edit sequences are competitive with GPT-4 and outperform models finetuned on the baseline dataset by +20% (+/-3%) in absolute score. Finally, we also pretrain our own tiny LMs for code understanding. We show that finetuning tiny models on synthetic code edits results in state-of-the-art code synthesis for the on-device model class. Our 150M parameter edit sequence LM matches or outperforms code models with twice as many parameters, both with and without repeated sampling, including Codex and AlphaCode.

Across medical specialties, clinical practice is anchored in evidence-based guidelines that codify best studied diagnostic and treatment pathways. These pathways routinely fall short for the long tail of real-world care not covered by guidelines. Most medical large language models (LLMs), however, are trained to encode common, guideline-focused medical knowledge in their parameters. Current evaluations test models primarily on recalling and reasoning with this memorized content, often in multiple-choice settings. Given the fundamental importance of evidence-based reasoning in medicine, it is neither feasible nor reliable to depend on memorization in practice. To address this gap, we introduce OGCaReBench, a free-form retrieval-focused benchmark aimed at evaluating LLMs at answering clinical questions that require going beyond typical guidelines. Extracted from published medical case reports and validated by medical experts, OGCaReBench contains long-form clinical questions requiring free-text answers, providing a systematic framework for assessing open-ended medical reasoning in rare, case-based scenarios. Our experiments reveal that even the best-performing baseline (GPT-5.2) correctly answers only 56% of our benchmark with specialized models only reaching 42%. Augmenting models with retrieved medical articles improves this performance to up to 82% (using GPT-5.2) highlighting the importance of evidence-grounding for real-world medical reasoning tasks. This work thus establishes a foundation for benchmarking and advancing both general-purpose and medical LLMs to produce reliable answers in challenging clinical contexts.

Recent research has leveraged large language model multi-agent systems for complex problem-solving while trying to reduce the manual effort required to build them, driving the development of automated agent workflow optimization methods. However, existing methods remain inflexible due to representational limitations, a lack of adaptability, and poor scalability when relying on discrete optimization techniques. We address these challenges with ScoreFlow, a simple yet high-performance framework that leverages efficient gradient-based optimization in a continuous space. ScoreFlow incorporates Score-DPO, a novel variant of the direct preference optimization method that accounts for quantitative feedback. Across six benchmarks spanning question answering, coding, and mathematical reasoning, ScoreFlow achieves an 8.2% improvement over existing baselines. Moreover, it empowers smaller models to outperform larger ones with lower inference costs. Project: https://github.com/Gen-Verse/ScoreFlow

A major technique in learning-augmented online algorithms is combining multiple algorithms or predictors. Since the performance of each predictor may vary over time, it is desirable to use not the single best predictor as a benchmark, but rather a dynamic combination which follows different predictors at different times. We design algorithms that combine predictions and are competitive against such dynamic combinations for a wide class of online problems, namely, metrical task systems. Against the best (in hindsight) unconstrained combination of ell predictors, we obtain a competitive ratio of O(ell^2), and show that this is best possible. However, for a benchmark with slightly constrained number of switches between different predictors, we can get a (1+epsilon)-competitive algorithm. Moreover, our algorithms can be adapted to access predictors in a bandit-like fashion, querying only one predictor at a time. An unexpected implication of one of our lower bounds is a new structural insight about covering formulations for the k-server problem.

Large-scale multilingual bitext often contains two distinct problems: non-parallel sentence pairs and low-quality translations. We decompose model-based assessment for such data into two independent components: parallelism assessment with multilingual embeddings and reference-free quality estimation (QE). For parallelism, we benchmark four embedding models on FLORES-200 and BOUQuET retrieval tasks, covering 6,654 source--target directions in our target language-pair inventory. For QE, we evaluate nine reference-free evaluators on professional FLORES-200 translations across 41,412 ordered source--target directions. Results show that no model is universally reliable across translation directions. Naive QE ensembles dilute strong model signals, while documented target-language coverage is strongly associated with higher QE scores. Overall, these findings suggest that multilingual parallel-data assessment is best approached as a direction-aware routing and calibration problem, where no single universal metric is expected to suffice across all languages.

The rapid proliferation of benchmarks for evaluating large language models (LLMs) has created an urgent need for systematic methods to assess benchmark quality itself. We propose Benchmark^2, a comprehensive framework comprising three complementary metrics: (1) Cross-Benchmark Ranking Consistency, measuring whether a benchmark produces model rankings aligned with peer benchmarks; (2) Discriminability Score, quantifying a benchmark's ability to differentiate between models; and (3) Capability Alignment Deviation, identifying problematic instances where stronger models fail but weaker models succeed within the same model family. We conduct extensive experiments across 15 benchmarks spanning mathematics, reasoning, and knowledge domains, evaluating 11 LLMs across four model families. Our analysis reveals significant quality variations among existing benchmarks and demonstrates that selective benchmark construction based on our metrics can achieve comparable evaluation performance with substantially reduced test sets.

Preference learning is critical for aligning large language models (LLMs) with human values, yet its success hinges on high-quality datasets comprising three core components: Preference Annotations, Instructions, and Response Pairs. Current approaches conflate these components, obscuring their individual impacts and hindering systematic optimization. In this work, we propose AIR, a component-wise analysis framework that systematically isolates and optimizes each component while evaluating their synergistic effects. Through rigorous experimentation, AIR reveals actionable principles: annotation simplicity (point-wise generative scoring), instruction inference stability (variance-based filtering across LLMs), and response pair quality (moderate margins + high absolute scores). When combined, these principles yield +5.3 average gains over baseline method, even with only 14k high-quality pairs. Our work shifts preference dataset design from ad hoc scaling to component-aware optimization, offering a blueprint for efficient, reproducible alignment.

We introduce a new dataset for Question Rewriting in Conversational Context (QReCC), which contains 14K conversations with 80K question-answer pairs. The task in QReCC is to find answers to conversational questions within a collection of 10M web pages (split into 54M passages). Answers to questions in the same conversation may be distributed across several web pages. QReCC provides annotations that allow us to train and evaluate individual subtasks of question rewriting, passage retrieval and reading comprehension required for the end-to-end conversational question answering (QA) task. We report the effectiveness of a strong baseline approach that combines the state-of-the-art model for question rewriting, and competitive models for open-domain QA. Our results set the first baseline for the QReCC dataset with F1 of 19.10, compared to the human upper bound of 75.45, indicating the difficulty of the setup and a large room for improvement.

Despite its simplicity and efficacy, the high token expenditure of self-consistency can limit its practical utility. Here we investigate if self-consistency can be made more token-efficient for long chain-of-thought reasoning tasks, while preserving its parallelism, through early hypothesis pruning. Concretely, we generate all solutions in parallel, but periodically prune intermediate hypotheses that are deemed unnecessary based on two lightweight indicators: (a) the model's own confidence in individual hypotheses, and (b) lexical coverage of all current hypotheses by candidate subsets that are under consideration for continued retention. We design a fast weighted set cover algorithm that utilizes the two indicators; our evaluation of five LLMs on three math benchmarks shows that this method can improve token efficiency for all models, by 10-35% in many cases.

Large language models are increasingly deployed as automated judges for evaluating other models, yet the stability of their verdicts under semantically equivalent prompt paraphrases remains unmeasured. We introduce JudgeSense, a framework and benchmark for quantifying this property via the Judge Sensitivity Score (JSS), defined as the fraction of paraphrase pairs on which a judge returns an identical decision. Evaluating nine judge models on 494 validated paraphrase pairs, we find that coherence is the only task where judges meaningfully differ, with JSS ranging from 0.389 to 0.992. On factuality, all judges cluster near JSS about 0.63, driven by a polarity-inverted prompt artifact; after correction, factuality JSS rises to about 0.9. Pairwise tasks (preference and relevance) exhibit degenerate always-A behavior in 8 of 9 judges, indicating strong position bias. Model scale does not predict consistency. We release code, decision logs, and a validated paraphrase dataset to support standardized JSS reporting.

We introduce BM25S, an efficient Python-based implementation of BM25 that only depends on Numpy and Scipy. BM25S achieves up to a 500x speedup compared to the most popular Python-based framework by eagerly computing BM25 scores during indexing and storing them into sparse matrices. It also achieves considerable speedups compared to highly optimized Java-based implementations, which are used by popular commercial products. Finally, BM25S reproduces the exact implementation of five BM25 variants based on Kamphuis et al. (2020) by extending eager scoring to non-sparse variants using a novel score shifting method. The code can be found at https://github.com/xhluca/bm25s

In this paper, we uncover a systematic bias in the evaluation paradigm of adopting large language models~(LLMs), e.g., GPT-4, as a referee to score and compare the quality of responses generated by candidate models. We find that the quality ranking of candidate responses can be easily hacked by simply altering their order of appearance in the context. This manipulation allows us to skew the evaluation result, making one model appear considerably superior to the other, e.g., Vicuna-13B could beat ChatGPT on 66 over 80 tested queries with ChatGPT as an evaluator. To address this issue, we propose a calibration framework with three simple yet effective strategies: 1) Multiple Evidence Calibration, which requires the evaluator model to generate multiple evaluation evidence before assigning ratings; 2) Balanced Position Calibration, which aggregates results across various orders to determine the final score; 3) Human-in-the-Loop Calibration, which introduces a balanced position diversity entropy to measure the difficulty of each example and seeks human assistance when needed. We also manually annotate the "win/tie/lose" outcomes of responses from ChatGPT and Vicuna-13B in the Vicuna Benchmark's question prompt, and extensive experiments demonstrate that our approach successfully mitigates evaluation bias, resulting in closer alignment with human judgments. We release our code and human annotation at https://github.com/i-Eval/FairEval to facilitate future research.

Self-consistency (SC), leveraging multiple samples from LLMs, shows significant gains on various reasoning tasks but struggles with free-form generation due to the difficulty of aggregating answers. Its variants, UCS and USC, rely on sample selection or voting mechanisms to improve output quality. These methods, however, face limitations due to their inability to fully utilize the nuanced consensus knowledge present within multiple candidate samples, often resulting in suboptimal outputs. We propose Fine-Grained Self-Consistency (FSC) to addresses these limitations by extracting and integrating segment-level commonalities from candidate samples, enhancing the performance of LLMs both in open-ended and reasoning tasks. Based on this, we present two additional strategies: candidate filtering, which enhances overall quality by identifying highly similar candidate sets, and merging, which reduces input token requirements by combining similar samples. The effectiveness of FSC is demonstrated through extensive experiments on various tasks, including summarization, code generation, and mathematical reasoning, using GPT-3.5-turbo and GPT-4. The results indicate significant improvements over baseline methods, showcasing the potential of FSC to optimize output quality by effectively synthesizing fine-grained consensus knowledge from multiple samples.

As Large Language Models (LLMs) saturate elementary benchmarks, the research frontier has shifted from generation to the reliability of automated evaluation. We demonstrate that standard "LLM-as-a-Judge" protocols suffer from a systematic Alignment Gap when applied to upper-undergraduate to early graduate level mathematics. To quantify this, we introduce QEDBench, the first large-scale dual-rubric alignment benchmark to systematically measure alignment with human experts on university-level math proofs by contrasting course-specific rubrics against expert common knowledge criteria. By deploying a dual-evaluation matrix (7 judges x 5 solvers) against 1,000+ hours of human evaluation, we reveal that certain frontier evaluators like Claude Opus 4.5, DeepSeek-V3, Qwen 2.5 Max, and Llama 4 Maverick exhibit significant positive bias (up to +0.18, +0.20, +0.30, +0.36 mean score inflation, respectively). Furthermore, we uncover a critical reasoning gap in the discrete domain: while Gemini 3.0 Pro achieves state-of-the-art performance (0.91 average human evaluation score), other reasoning models like GPT-5 Pro and Claude Sonnet 4.5 see their performance significantly degrade in discrete domains. Specifically, their average human evaluation scores drop to 0.72 and 0.63 in Discrete Math, and to 0.74 and 0.50 in Graph Theory. In addition to these research results, we also release QEDBench as a public benchmark for evaluating and improving AI judges. Our benchmark is publicly published at https://github.com/qqliu/Yale-QEDBench.

Despite their widespread applications, Large Language Models (LLMs) often struggle to express uncertainty, posing a challenge for reliable deployment in high stakes and safety critical domains like clinical diagnostics. Existing standard baseline methods such as model logits and elicited probabilities produce overconfident and poorly calibrated estimates. In this work, we propose Approximate Bayesian Computation (ABC), a likelihood-free Bayesian inference, based approach that treats LLMs as a stochastic simulator to infer posterior distributions over predictive probabilities. We evaluate our ABC approach on two clinically relevant benchmarks: a synthetic oral lesion diagnosis dataset and the publicly available GretelAI symptom-to-diagnosis dataset. Compared to standard baselines, our approach improves accuracy by up to 46.9\%, reduces Brier scores by 74.4\%, and enhances calibration as measured by Expected Calibration Error (ECE) and predictive entropy.

We introduce Metric-Fair Prompting, a fairness-aware prompting framework that guides large language models (LLMs) to make decisions under metric-fairness constraints. In the application of multiple-choice medical question answering, each {(question, option)} pair is treated as a binary instance with label +1 (correct) or -1 (incorrect). To promote {individual fairness}~--~treating similar instances similarly~--~we compute question similarity using NLP embeddings and solve items in joint pairs of similar questions rather than in isolation. The prompt enforces a global decision protocol: extract decisive clinical features, map each \((question, option)\) to a score f(x) that acts as confidence, and impose a Lipschitz-style constraint so that similar inputs receive similar scores and, hence, consistent outputs. Evaluated on the {MedQA (US)} benchmark, Metric-Fair Prompting is shown to improve performance over standard single-item prompting, demonstrating that fairness-guided, confidence-oriented reasoning can enhance LLM accuracy on high-stakes clinical multiple-choice questions.

Rubric-based evaluation has become a prevailing paradigm for evaluating instruction following in large language models (LLMs). Despite its widespread use, the reliability of these rubric-level evaluations remains unclear, calling for meta-evaluation. However, prior meta-evaluation efforts largely focus on the response level, failing to assess the fine-grained judgment accuracy that rubric-based evaluation relies on. To bridge this gap, we introduce RubricEval. Our benchmark features: (1) the first rubric-level meta-evaluation benchmark for instruction following, (2) diverse instructions and responses spanning multiple categories and model sources, and (3) a substantial set of 3,486 quality-controlled instances, along with Easy/Hard subsets that better differentiates judge performance. Our experiments reveal that rubric-level judging remains far from solved: even GPT-4o, a widely adopted judge in instruction-following benchmarks, achieves only 55.97% on Hard subset. Considering evaluation paradigm, rubric-level evaluation outperforms checklist-level, explicit reasoning improves accuracy, and both together reduce inter-judge variance. Through our established rubric taxonomy, we further identify common failure modes and offer actionable insights for reliable instruction-following evaluation.

Best-of-N (BoN) sampling, a common strategy for test-time scaling of Large Language Models (LLMs), relies on reward models to select the best candidate solution from multiple generations. However, traditional reward models often assign arbitrary and inconsistent scores, limiting their effectiveness. To address this, we propose a Pairwise Reward Model (Pairwise RM) combined with a knockout tournament for BoN sampling. Instead of assigning absolute scores, given one math problem, Pairwise RM evaluates two candidate solutions' correctness simultaneously. This approach eliminates the need for arbitrary scoring and enables cross-validation of solutions through parallel comparison. In the knockout tournament, Pairwise RM conducts pairwise comparisons between candidate solutions and eliminates the incorrect ones iteratively. We construct \ourdataset, a large-scale dataset of 443K pairwise comparisons derived from NumiaMath and annotated using gemini-1.5-flash, and train the Pairwise RM via supervised fine-tuning. Experiments on MATH-500 and the Olympiad Bench demonstrate significant improvements over traditional discriminative reward models. And a 40\% to 60\% relative improvement is achieved on the top 50\% challenging problems.

The training data for fine-tuning large language models (LLMs) is typically structured as input-output pairs. However, for many tasks, there can be multiple equally valid output variations for the same input. Recent studies have observed that the choice of output variation used in training can affect the model's performance. This raises an important question: how can we generate the most effective output from the many possible response generation strategy options? Rather than relying on the traditional but resource-intensive train-and-evaluate approach, this paper proposes a scalable, approximate method for estimating the quality of a small subset of generated training data derived from the same input. We then evaluate how well this small subset of generated output fits the target model we are trying to train. We present a large-scale benchmark covering diverse reasoning-based datasets to support our study. The central idea is that a good output should closely resemble the output generated by the target LLM. We formalize this 'closeness' as the expected alignment score between a candidate output and the output sampled from the target LLM. We connect this measurement to the perplexity metric used in previous literature and demonstrate that leveraging an alignment-based metric can provide better predictions of model performance. Using this strategy, we can evaluate a small subset of the generated output from each response generation strategy option, then select the most effective strategy. We show that an LLM trained on data generated by the selected strategy could lead to a significant performance gain in many cases.

LLM agents hold significant promise for advancing scientific research. To accelerate this progress, we introduce AIRS-Bench (the AI Research Science Benchmark), a suite of 20 tasks sourced from state-of-the-art machine learning papers. These tasks span diverse domains, including language modeling, mathematics, bioinformatics, and time series forecasting. AIRS-Bench tasks assess agentic capabilities over the full research lifecycle -- including idea generation, experiment analysis and iterative refinement -- without providing baseline code. The AIRS-Bench task format is versatile, enabling easy integration of new tasks and rigorous comparison across different agentic frameworks. We establish baselines using frontier models paired with both sequential and parallel scaffolds. Our results show that agents exceed human SOTA in four tasks but fail to match it in sixteen others. Even when agents surpass human benchmarks, they do not reach the theoretical performance ceiling for the underlying tasks. These findings indicate that AIRS-Bench is far from saturated and offers substantial room for improvement. We open-source the AIRS-Bench task definitions and evaluation code to catalyze further development in autonomous scientific research.

This research introduces the Judge's Verdict Benchmark, a novel two-step methodology to evaluate Large Language Models (LLMs) as judges for response accuracy evaluation tasks. We assess how well 54 LLMs can replicate human judgment when scoring responses from RAG (Retrieval-Augmented Generation) or Agentic pipelines against ground truth answers. Our methodology progresses from traditional correlation analysis to comprehensive Cohen's Kappa analysis that measures actual agreement patterns. The two-step approach includes: (1) a correlation test that filters judges with strong alignment, followed by (2) a human-likeness test using z-scores to identify two distinct judgment patterns: human-like judgment (|z| < 1) that mimics natural human variation, and super-consistent judgment (z > 1) that exceeds typical human-to-human agreement levels. This methodology reveals that 27 out of 54 tested LLMs achieve Tier 1 performance: 23 models exhibit human-like patterns that preserve the nuances of human judgment, while 4 models demonstrate super-consistent behavior, a pattern that could indicate either enhanced reliability or oversimplification of complex judgments. Testing 43 open-source models (1B-405B parameters) and 11 closed models (GPT, Gemini, Claude variants), we demonstrate that judge excellence is not solely dependent on model size but on specific training strategies. Our key contributions include: (1) establishing that correlation alone is insufficient for judge evaluation, (2) introducing a "Turing Test for judges" based on agreement patterns, and (3) providing a standardized benchmark for classifying LLM judges into distinct performance tiers for different evaluation needs.

Self-consistency improves large language models by sampling multiple reasoning paths and selecting the most frequent answer, but majority voting often fails to recover correct answers that are already present among the samples. We address this limitation with Ranking-Improved Self-Consistency (RISC), which reformulates answer selection in self-consistency as a ranking problem. Instead of relying on a single uncertainty or confidence signal, RISC uses a lightweight LambdaRank model to score candidate answers with five carefully designed features that capture answer frequency, semantic centrality, and reasoning-trace consistency. We evaluate RISC on three datasets under a range of test-time budgets. Across datasets, RISC consistently achieves a better accuracy-efficiency trade-off than standard self-consistency and strong baselines, with particularly large gains on question answering benchmarks. Further analysis shows that the proposed features are individually useful and, more importantly, complementary, highlighting the value of learning to combine multiple informative signals for test-time answer selection.

Test-time scaling evaluates reasoning LLMs by sampling multiple outputs per prompt, but ranking models in this regime remains underexplored. We formalize dense benchmark ranking under test-time scaling and introduce Scorio, a library that implements statistical ranking methods such as paired-comparison models, item response theory (IRT) models, voting rules, and graph- and spectral-based methods. Across 20 reasoning models on four Olympiad-style math benchmarks (AIME'24, AIME'25, HMMT'25, and BrUMO'25; up to N=80 trials), most full-trial rankings agree closely with the Bayesian gold standard Bayes_{U}@80 (mean Kendall's τ_b = 0.93--0.95), and 19--34 methods recover exactly the same ordering. In the single-trial regime, the best methods reach τ_b approx 0.86. Using greedy decoding as an empirical prior (Bayes_{R_0}@N) reduces variance at N=1 by 16--52%, but can bias rankings when greedy and stochastic sampling disagree. These results identify reliable ranking methods for both high- and low-budget test-time scaling. We release Scorio as an open-source library at https://github.com/mohsenhariri/scorio.

Pre-trained LLMs have demonstrated substantial capabilities across a range of conventional natural language processing (NLP) tasks, such as summarization and entity recognition. In this paper, we explore the application of LLMs in the generation of high-quality protein sequences. Specifically, we adopt a suite of pre-trained LLMs, including Mistral-7B1, Llama-2-7B2, Llama-3-8B3, and gemma-7B4, to produce valid protein sequences. All of these models are publicly available.5 Unlike previous work in this field, our approach utilizes a relatively small dataset comprising 42,000 distinct human protein sequences. We retrain these models to process protein-related data, ensuring the generation of biologically feasible protein structures. Our findings demonstrate that even with limited data, the adapted models exhibit efficiency comparable to established protein-focused models such as ProGen varieties, ProtGPT2, and ProLLaMA, which were trained on millions of protein sequences. To validate and quantify the performance of our models, we conduct comparative analyses employing standard metrics such as pLDDT, RMSD, TM-score, and REU. Furthermore, we commit to making the trained versions of all four models publicly available, fostering greater transparency and collaboration in the field of computational biology.

Retrieval Augmented Generation (RAG) has become one of the most popular paradigms for enabling LLMs to access external data, and also as a mechanism for grounding to mitigate against hallucinations. When implementing RAG you can face several challenges like effective integration of retrieval models, efficient representation learning, data diversity, computational efficiency optimization, evaluation, and quality of text generation. Given all these challenges, every day a new technique to improve RAG appears, making it unfeasible to experiment with all combinations for your problem. In this context, this paper presents good practices to implement, optimize, and evaluate RAG for the Brazilian Portuguese language, focusing on the establishment of a simple pipeline for inference and experiments. We explored a diverse set of methods to answer questions about the first Harry Potter book. To generate the answers we used the OpenAI's gpt-4, gpt-4-1106-preview, gpt-3.5-turbo-1106, and Google's Gemini Pro. Focusing on the quality of the retriever, our approach achieved an improvement of MRR@10 by 35.4% compared to the baseline. When optimizing the input size in the application, we observed that it is possible to further enhance it by 2.4%. Finally, we present the complete architecture of the RAG with our recommendations. As result, we moved from a baseline of 57.88% to a maximum relative score of 98.61%.

Multi-modal generative document parsing systems challenge traditional evaluation: unlike deterministic OCR or layout models, they often produce semantically correct yet structurally divergent outputs. Conventional metrics-CER, WER, IoU, or TEDS-misclassify such diversity as error, penalizing valid interpretations and obscuring system behavior. We introduce SCORE (Structural and COntent Robust Evaluation), an interpretation-agnostic framework that integrates (i) adjusted edit distance for robust content fidelity, (ii) token-level diagnostics to distinguish hallucinations from omissions, (iii) table evaluation with spatial tolerance and semantic alignment, and (iv) hierarchy-aware consistency checks. Together, these dimensions enable evaluation that embraces representational diversity while enforcing semantic rigor. Across 1,114 pages spanning a holistic benchmark and a field dataset, SCORE consistently revealed cross-dataset performance patterns missed by standard metrics. In 2-5% of pages with ambiguous table structures, traditional metrics penalized systems by 12-25% on average, leading to distorted rankings. SCORE corrected these cases, recovering equivalence between alternative but valid interpretations. Moreover, by normalizing generative outputs into a format-agnostic representation, SCORE reproduces traditional scores (e.g., table F1 up to 0.93) without requiring object-detection pipelines, demonstrating that generative parsing alone suffices for comprehensive evaluation. By exposing how interpretive diversity impacts evaluation outcomes and providing multi-dimensional, interpretable diagnostics, SCORE establishes foundational principles for semantically grounded, fair, and practical benchmarking of modern document parsing systems.

Anomaly detection (AD) is a machine learning task that identifies anomalies by learning patterns from normal training data. In many real-world scenarios, anomalies vary in severity, from minor anomalies with little risk to severe abnormalities requiring immediate attention. However, existing models primarily operate in a binary setting, and the anomaly scores they produce are usually based on the deviation of data points from normal data, which may not accurately reflect practical severity. In this paper, we address this gap by making three key contributions. First, we propose a novel setting, Multilevel AD (MAD), in which the anomaly score represents the severity of anomalies in real-world applications, and we highlight its diverse applications across various domains. Second, we introduce a novel benchmark, MAD-Bench, that evaluates models not only on their ability to detect anomalies, but also on how effectively their anomaly scores reflect severity. This benchmark incorporates multiple types of baselines and real-world applications involving severity. Finally, we conduct a comprehensive performance analysis on MAD-Bench. We evaluate models on their ability to assign severity-aligned scores, investigate the correspondence between their performance on binary and multilevel detection, and study their robustness. This analysis offers key insights into improving AD models for practical severity alignment. The code framework and datasets used for the benchmark will be made publicly available.

Large language models (LLMs) are now widely used to evaluate the quality of text, a field commonly referred to as LLM-as-a-judge. While prior works mainly focus on point-wise and pair-wise evaluation paradigms. Rubric-based evaluation, where LLMs select a score from multiple rubrics, has received less analysis. In this work, we show that rubric-based evaluation implicitly resembles a multi-choice setting and therefore has position bias: LLMs prefer score options appearing at specific positions in the rubric list. Through controlled experiments across multiple models and datasets, we demonstrate consistent position bias. To mitigate this bias, we propose a balanced permutation strategy that evenly distributes each score option across positions. We show that aggregating scores across balanced permutations not only reveals latent position bias, but also improves correlation between the LLM-as-a-Judge and human. Our results suggest that rubric-based LLM-as-a-Judge is not inherently point-wise and that simple permutation-based calibration can substantially improve its reliability.

Nucleotide sequences constitute the fundamental genetic basis of biological systems, rendering viral genomic analysis critical for biomedical advancement. Despite progress in biological foundation models, specifically nucleotide foundation models (NFMs), the field lacks a unified standard for viral genomics to facilitate community development and enforce biosecurity constraints. To address this, we introduce ViroBench, the first comprehensive and large-scale benchmark specifically designed for NFMs in viral settings. ViroBench evaluates models across two critical dimensions: biological understanding and latent biosecurity risk, covering 18 diverse scenarios within 4 task types. Extensive evaluation of 66 NFMs across diverse architectures yields three critical conclusions. Firstly, NFMs exhibit a performance degradation in biological understanding under phylogenetic and temporal shifts, indicating weak extrapolation capabilities. Secondly, generation tasks reveal a decoupling between statistical likelihood and biological functional validity, posing latent biosecurity risks. Thirdly, controlled ablation studies reveal that taxonomic diversity in pretraining data outweighs parameter scale. Specifically, a lightweight baseline trained on diverse data achieves a 67.5% performance gain over its original model. Overall, ViroBench provides interpretable, diagnostic evaluations and a reproducible measurement framework for future research on viral nucleotide foundation models. The datasets and code are publicly available at https://github.com/QIANJINYDX/ViroBench.

This paper describes a system designed to distinguish between AI-generated and human-written scientific excerpts in the DAGPap24 competition hosted within the Fourth Workshop on Scientific Document Processing. In this competition the task is to find artificially generated token-level text fragments in documents of a scientific domain. Our work focuses on the use of a multi-task learning architecture with two heads. The application of this approach is justified by the specificity of the task, where class spans are continuous over several hundred characters. We considered different encoder variations to obtain a state vector for each token in the sequence, as well as a variation in splitting fragments into tokens to further feed into the input of a transform-based encoder. This approach allows us to achieve a 9% quality improvement relative to the baseline solution score on the development set (from 0.86 to 0.95) using the average macro F1-score, as well as a score of 0.96 on a closed test part of the dataset from the competition.

Large language models achieve high task performance yet often hallucinate or rely on outdated knowledge. Retrieval-augmented generation (RAG) addresses these gaps by coupling generation with external search. We analyse how hyperparameters influence speed and quality in RAG systems, covering Chroma and Faiss vector stores, chunking policies, cross-encoder re-ranking, and temperature, and we evaluate six metrics: faithfulness, answer correctness, answer relevancy, context precision, context recall, and answer similarity. Chroma processes queries 13% faster, whereas Faiss yields higher retrieval precision, revealing a clear speed-accuracy trade-off. Naive fixed-length chunking with small windows and minimal overlap outperforms semantic segmentation while remaining the quickest option. Re-ranking provides modest gains in retrieval quality yet increases runtime by roughly a factor of 5, so its usefulness depends on latency constraints. These results help practitioners balance computational cost and accuracy when tuning RAG systems for transparent, up-to-date responses. Finally, we re-evaluate the top configurations with a corrective RAG workflow and show that their advantages persist when the model can iteratively request additional evidence. We obtain a near-perfect context precision (99%), which demonstrates that RAG systems can achieve extremely high retrieval accuracy with the right combination of hyperparameters, with significant implications for applications where retrieval quality directly impacts downstream task performance, such as clinical decision support in healthcare.

Identifying drug-target interactions is essential for developing effective therapeutics. Binding affinity quantifies these interactions, and traditional approaches rely on computationally intensive 3D structural data. In contrast, language models can efficiently process sequential data, offering an alternative approach to molecular representation. In the current study, we introduce BAPULM, an innovative sequence-based framework that leverages the chemical latent representations of proteins via ProtT5-XL-U50 and ligands through MolFormer, eliminating reliance on complex 3D configurations. Our approach was validated extensively on benchmark datasets, achieving scoring power (R) values of 0.925 pm 0.043, 0.914 pm 0.004, and 0.8132 pm 0.001 on benchmark1k2101, Test2016_290, and CSAR-HiQ_36, respectively. These findings indicate the robustness and accuracy of BAPULM across diverse datasets and underscore the potential of sequence-based models in-silico drug discovery, offering a scalable alternative to 3D-centric methods for screening potential ligands.

This paper introduces BioAgent Bench, a benchmark dataset and an evaluation suite designed for measuring the performance and robustness of AI agents in common bioinformatics tasks. The benchmark contains curated end-to-end tasks (e.g., RNA-seq, variant calling, metagenomics) with prompts that specify concrete output artifacts to support automated assessment, including stress testing under controlled perturbations. We evaluate frontier closed-source and open-weight models across multiple agent harnesses, and use an LLM-based grader to score pipeline progress and outcome validity. We find that frontier agents can complete multi-step bioinformatics pipelines without elaborate custom scaffolding, often producing the requested final artifacts reliably. However, robustness tests reveal failure modes under controlled perturbations (corrupted inputs, decoy files, and prompt bloat), indicating that correct high-level pipeline construction does not guarantee reliable step-level reasoning. Finally, because bioinformatics workflows may involve sensitive patient data, proprietary references, or unpublished IP, closed-source models can be unsuitable under strict privacy constraints; in such settings, open-weight models may be preferable despite lower completion rates. We release the dataset and evaluation suite publicly.

Equivariance theory predicts that an architectural symmetry prior reduces sample complexity by a factor of |G|; this is widely cited but rarely measured as a scaling law with controls that separate the prior from its confounds. On a controlled C_n-symmetric task, we report three findings. First, a wrong-group control with identical orbit size and matched compute is worse than no constraint (joint pairwise CI [+0.79, +3.26] excludes zero, robust across estimators); misaligned constraint is actively harmful, not merely unhelpful. Second, an augmentation baseline equipped with test-time orbit averaging matches the equivariant model exactly -- bit-identical per-epoch validation curves across matched cells -- so the architecture-vs-augmentation gap is conditional on asymmetric test-time computation, not unconditional. Third, the relative exchange rate beta_diff = 1.28 is consistent in sign and order of magnitude with the theoretical 1.0 (single-level CI [+0.92, +2.05]); the more conservative two-level bootstrap (seeds x group sizes) widens this to [-0.63, +1.72], including zero, and a finer-N replication on a sqrt(2)-spaced grid is inconclusive (point estimate -0.82). The methodological contributions -- the relative-rate estimator that cancels the shared-difficulty confound, the wrong-group control, and a pre-specified failure taxonomy -- transfer to any inductive bias whose strength can be parameterised. Honest scoping: the primary estimator beta_diff was adopted post-hoc after the initial analysis revealed a positive-slope identifiability problem; the design was never externally pre-registered; and the headline number rests on an OLS slope over seven group sizes on a coarse N grid. This is an exploratory study, not a confirmatory measurement; the wrong-group result is the cleanest finding and the one we report with the most confidence. A registered replication on fresh seeds is future work.

We introduce a novel framework for measuring Overton pluralism in LLMs--the extent to which diverse viewpoints are represented in model outputs. We (i) formalize Overton pluralism as a set coverage metric (OvertonScore), (ii) conduct a large-scale U.S.-representative human study (N = 1209; 60 questions; 8 LLMs), and (iii) develop an automated benchmark that closely reproduces human judgments. On average, models achieve OvertonScores of 0.35--0.41, with DeepSeek V3 performing best; yet all models remain far below the theoretical maximum of 1.0, revealing substantial headroom for improvement. Because repeated large-scale human studies are costly and slow, scalable evaluation tools are essential for model development. Hence, we propose an automated benchmark that achieves high rank correlation with human judgments (ρ=0.88), providing a practical proxy without replacing human assessment. By turning pluralistic alignment from a normative aim into a measurable benchmark, our work establishes a foundation for systematic progress toward more pluralistic LLMs.

Peer reviewing is a central component in the scientific publishing process. We present the first public dataset of scientific peer reviews available for research purposes (PeerRead v1) providing an opportunity to study this important artifact. The dataset consists of 14.7K paper drafts and the corresponding accept/reject decisions in top-tier venues including ACL, NIPS and ICLR. The dataset also includes 10.7K textual peer reviews written by experts for a subset of the papers. We describe the data collection process and report interesting observed phenomena in the peer reviews. We also propose two novel NLP tasks based on this dataset and provide simple baseline models. In the first task, we show that simple models can predict whether a paper is accepted with up to 21% error reduction compared to the majority baseline. In the second task, we predict the numerical scores of review aspects and show that simple models can outperform the mean baseline for aspects with high variance such as 'originality' and 'impact'.

Cloze testing is a common method for measuring the behavior of large language models on a number of benchmark tasks. Using the MMLU dataset, we show that the base-rate probability (BRP) differences across answer tokens are significant and affect task performance ie. guess A if uncertain. We find that counterfactual prompting does sufficiently mitigate the BRP effect. The BRP effect is found to have a similar effect to test taking strategies employed by humans leading to the conflation of task performance and test-taking ability. We propose the Nvr-X-MMLU task, a variation of MMLU, which helps to disambiguate test-taking ability from task performance and reports the latter.

Retrieval-Augmented Generation (RAG) systems face challenges with complex, multihop questions, and agentic frameworks such as Search-R1 (Jin et al., 2025), which operates iteratively, have been proposed to address these complexities. However, such approaches can introduce inefficiencies, including repetitive retrieval of previously processed information and challenges in contextualizing retrieved results effectively within the current generation prompt. Such issues can lead to unnecessary retrieval turns, suboptimal reasoning, inaccurate answers, and increased token consumption. In this paper, we investigate test-time modifications to the Search-R1 pipeline to mitigate these identified shortcomings. Specifically, we explore the integration of two components and their combination: a contextualization module to better integrate relevant information from retrieved documents into reasoning, and a de-duplication module that replaces previously retrieved documents with the next most relevant ones. We evaluate our approaches using the HotpotQA (Yang et al., 2018) and the Natural Questions (Kwiatkowski et al., 2019) datasets, reporting the exact match (EM) score, an LLM-as-a-Judge assessment of answer correctness, and the average number of turns. Our best-performing variant, utilizing GPT-4.1-mini for contextualization, achieves a 5.6% increase in EM score and reduces the number of turns by 10.5% compared to the Search-R1 baseline, demonstrating improved answer accuracy and retrieval efficiency.

Increasing test-time computation has emerged as a promising direction for improving language model performance, particularly in scenarios where model finetuning is impractical or impossible due to computational constraints or private model weights. However, existing test-time search methods using a reward model (RM) often degrade in quality as compute scales, due to the over-optimization of what are inherently imperfect reward proxies. We introduce QAlign, a new test-time alignment approach. As we scale test-time compute, QAlign converges to sampling from the optimal aligned distribution for each individual prompt. By adopting recent advances in Markov chain Monte Carlo for text generation, our method enables better-aligned outputs without modifying the underlying model or even requiring logit access. We demonstrate the effectiveness of QAlign on mathematical reasoning benchmarks (GSM8K and GSM-Symbolic) using a task-specific RM, showing consistent improvements over existing test-time compute methods like best-of-n and majority voting. Furthermore, when applied with more realistic RMs trained on the Tulu 3 preference dataset, QAlign outperforms direct preference optimization (DPO), best-of-n, majority voting, and weighted majority voting on a diverse range of datasets (GSM8K, MATH500, IFEval, MMLU-Redux, and TruthfulQA). A practical solution to aligning language models at test time using additional computation without degradation, our approach expands the limits of the capability that can be obtained from off-the-shelf language models without further training.

Recently, using a powerful proprietary Large Language Model (LLM) (e.g., GPT-4) as an evaluator for long-form responses has become the de facto standard. However, for practitioners with large-scale evaluation tasks and custom criteria in consideration (e.g., child-readability), using proprietary LLMs as an evaluator is unreliable due to the closed-source nature, uncontrolled versioning, and prohibitive costs. In this work, we propose Prometheus, a fully open-source LLM that is on par with GPT-4's evaluation capabilities when the appropriate reference materials (reference answer, score rubric) are accompanied. We first construct the Feedback Collection, a new dataset that consists of 1K fine-grained score rubrics, 20K instructions, and 100K responses and language feedback generated by GPT-4. Using the Feedback Collection, we train Prometheus, a 13B evaluator LLM that can assess any given long-form text based on customized score rubric provided by the user. Experimental results show that Prometheus scores a Pearson correlation of 0.897 with human evaluators when evaluating with 45 customized score rubrics, which is on par with GPT-4 (0.882), and greatly outperforms ChatGPT (0.392). Furthermore, measuring correlation with GPT-4 with 1222 customized score rubrics across four benchmarks (MT Bench, Vicuna Bench, Feedback Bench, Flask Eval) shows similar trends, bolstering Prometheus's capability as an evaluator LLM. Lastly, Prometheus achieves the highest accuracy on two human preference benchmarks (HHH Alignment & MT Bench Human Judgment) compared to open-sourced reward models explicitly trained on human preference datasets, highlighting its potential as an universal reward model. We open-source our code, dataset, and model at https://github.com/kaistAI/Prometheus.

The automatic evaluation of instruction following typically involves using large language models (LLMs) to assess response quality. However, there is a lack of comprehensive evaluation of these LLM-based evaluators across two dimensions: the base LLMs and the evaluation protocols. Therefore, we present a thorough meta-evaluation of instruction following, including 25 base LLMs and 15 recently proposed evaluation protocols, on 4 human-annotated datasets, assessing the evaluation accuracy of the LLM-evaluators. Our evaluation allows us to identify the best-performing base LLMs and evaluation protocols with a high degree of robustness. Moreover, our large-scale evaluation reveals: (1) Base LLM performance ranking remains largely consistent across evaluation protocols, with less capable LLMs showing greater improvement from protocol enhancements; (2) Robust evaluation of evaluation protocols requires many base LLMs with varying capability levels, as protocol effectiveness can depend on the base LLM used; (3) Evaluation results on different datasets are not always consistent, so a rigorous evaluation requires multiple datasets with distinctive features. We release our meta-evaluation suite ReIFE, which provides the codebase and evaluation result collection for more than 500 LLM-evaluator configurations, to support future research in instruction-following evaluation.

The Abstraction and Reasoning Corpus (ARC-AGI) poses a significant challenge for large language models (LLMs), exposing limitations in their abstract reasoning abilities. In this work, we leverage task-specific data augmentations throughout the training, generation, and scoring phases, and employ a depth-first search algorithm to generate diverse, high-probability candidate solutions. Furthermore, we utilize the LLM not only as a generator but also as a scorer, using its output probabilities to select the most promising solutions. Our method achieves a score of 71.6% (286.5/400 solved tasks) on the public ARC-AGI evaluation set, demonstrating state-of-the-art performance among publicly available approaches. While concurrent closed-source work has reported higher scores, our method distinguishes itself through its transparency, reproducibility, and remarkably low inference cost, averaging only around 2ct per task on readily available hardware (we assume a price of 36ct/hour for a Nvidia 4090 GPU).

We present a cross-architecture evaluation of production LLM inference on AMD Instinct MI325X GPUs, benchmarking four models spanning 235B to 1 trillion parameters across three architectural families (MoE+MLA, Dense+GQA, MoE+GQA) on an 8-GPU cluster with 2TB aggregate HBM3e using vLLM v0.14.1. Our results demonstrate that architecture-aware optimization is essential: MLA models require block size 1 and cannot use KV cache offloading, while GQA models benefit from both. The AMD AITER runtime is required for competitive MLA inference throughput and must be selectively disabled for architectures with incompatible attention head configurations. A controlled AITER ablation on Llama-3.1-405B (n=5 per condition) reveals a modest 3-5% throughput benefit at high concurrency but 2-16x higher measurement variability, confirming that AITER's large speedups target MoE/MLA kernels specifically. Under text-only workloads, Llama-405B and DeepSeek V3.2 achieve comparable peak throughput (15,944 and 15,343 tok/s) despite an order-of-magnitude difference in active parameters. Under vision workloads, Qwen3-VL-235B reaches 47,873 tok/s, 6.5x higher than Kimi-K2.5 (7,327 tok/s). Active parameter count per token is associated with inference throughput, though confounded by differences in quantization, AITER acceleration, and tensor parallelism. All four models exhibit a common throughput saturation point consistent with a memory-bandwidth bottleneck (~500 concurrent for short sequences, ~100-200 for longer sequences). All models maintain 100% HTTP-level success rates through 1,000 concurrent users, processing 18.9 million tokens across 17,406 requests without failures.

Multimodal large language models (MLLMs) have advanced clinical tasks for common conditions, but their performance on rare diseases remains largely untested. In rare-disease scenarios, clinicians often lack prior clinical knowledge, forcing them to rely strictly on case-level evidence for clinical judgments. Existing benchmarks predominantly evaluate common-condition, single-image settings, leaving multimodal and multi-image evidence integration under rare-disease data scarcity systematically unevaluated. We introduce MMRareBench, to our knowledge the first rare-disease benchmark jointly evaluating multimodal and multi-image clinical capability across four workflow-aligned tracks: diagnosis, treatment planning, cross-image evidence alignment, and examination suggestion. The benchmark comprises 1,756 question-answer pairs with 7,958 associated medical images curated from PMC case reports, with Orphanet-anchored ontology alignment, track-specific leakage control, evidence-grounded annotations, and a two-level evaluation protocol. A systematic evaluation of 23 MLLMs reveals fragmented capability profiles and universally low treatment-planning performance, with medical-domain models trailing general-purpose MLLMs substantially on multi-image tracks despite competitive diagnostic scores. These patterns are consistent with a capacity dilution effect: medical fine-tuning can narrow the diagnostic gap but may erode the compositional multi-image capability that rare-disease evidence integration demands.

Large language models (LLMs) benefit from test-time scaling but are often hampered by high inference latency. Speculative decoding is a natural way to accelerate the scaling process; however, scaling along both the parallel and sequential dimensions poses significant challenges, including substantial memory-bound execution and synchronization overhead. We introduce ATTS (Asynchronous Test-Time Scaling), a statistically guaranteed adaptive scaling framework that follows the hypothesis testing process to address these challenges. By revisiting arithmetic intensity, ATTS identifies synchronization as the primary bottleneck. It enables asynchronous inference through online calibration and proposes an ordinal classification algorithm that supports a three-stage rejection sampling pipeline, scaling along both the sequential and parallel axes. Across experiments on the MATH, AMC23, AIME24, and AIME25 datasets and across multiple draft-target model families, we show that ATTS delivers up to 56.7x speedup in test-time scaling and a 4.14x throughput improvement, while maintaining accurate control of the rejection rate, reducing latency and memory overhead, and incurring no accuracy loss. By scaling both in parallel and sequential dimensions, we enable the 1.5B/70B draft/target model combination to achieve the performance of the state-of-the-art reasoning model o3-mini (high) on the AIME dataset. We have released the code at https://github.com/menik1126/asynchronous-test-time-scaling.

Large Language Models (LLMs) are commonly used as evaluators in various applications, but the reliability of the outcomes remains a challenge. One such challenge is using LLMs-as-judges for direct assessment, i.e., assigning scores from a specified range without any references. We first show that this challenge stems from LLM judge outputs being associated with score range bias, i.e., LLM judge outputs are highly sensitive to pre-defined score ranges, preventing the search for optimal score ranges. We also show that similar biases exist among models from the same family. We then mitigate this bias through contrastive decoding, achieving up to 11.3% relative improvement on average in Spearman correlation with human judgments across different score ranges.

Combination therapies have become the standard of care for diseases such as cancer, tuberculosis, malaria and HIV. However, the combinatorial set of available multi-drug treatments creates a challenge in identifying effective combination therapies available in a situation. To assist medical professionals in identifying beneficial drug-combinations, we construct an expert-annotated dataset for extracting information about the efficacy of drug combinations from the scientific literature. Beyond its practical utility, the dataset also presents a unique NLP challenge, as the first relation extraction dataset consisting of variable-length relations. Furthermore, the relations in this dataset predominantly require language understanding beyond the sentence level, adding to the challenge of this task. We provide a promising baseline model and identify clear areas for further improvement. We release our dataset, code, and baseline models publicly to encourage the NLP community to participate in this task.

Instruction-following is a foundational capability of large language models (LLMs), with its improvement hinging on scalable and accurate feedback from judge models. However, the reliability of current judge models in instruction-following remains underexplored due to several deficiencies of existing meta-evaluation benchmarks, such as their insufficient data coverage and oversimplified pairwise evaluation paradigms that misalign with model optimization scenarios. To this end, we propose IF-RewardBench, a comprehensive meta-evaluation benchmark for instruction-following that covers diverse instruction and constraint types. For each instruction, we construct a preference graph containing all pairwise preferences among multiple responses based on instruction-following quality. This design enables a listwise evaluation paradigm that assesses the capabilities of judge models to rank multiple responses, which is essential in guiding model alignment. Extensive experiments on IF-RewardBench reveal significant deficiencies in current judge models and demonstrate that our benchmark achieves a stronger positive correlation with downstream task performance compared to existing benchmarks. Our codes and data are available at https://github.com/thu-coai/IF-RewardBench.

Pre-training data mixtures are commonly tuned by running small-scale experiments and extrapolating to the target training budget. When high-quality data is scarce and must be repeated, this extrapolation frequently fails, but the source of the failure has not been isolated. We show that a primary culprit is a repetition mismatch: because high-quality datasets are small, their repetition rate changes as the training budget grows, shifting the optimal mixture in ways that small-scale proxy experiments do not anticipate. A subsampling procedure that matches the target repetition rate controls for this effect. In a two-source setting combining limited high-quality data with web crawl, a single repetition-controlled experiment using only 1/16 of the target tokens recovers a mixture within 0.05 of the optimum for a 757M parameter model, compared to an error of 0.75 without repetition control. Achieving comparable accuracy without repetition control requires three to four horizons, consuming 44 to 94% of the target token budget. With three data sources, the larger mixture space requires more than a single experiment to constrain, but the approach remains effective: at the 757M scale, just two repetition-controlled horizons recover the optimal mixture, outperforming baselines that instead require the full two-source experiments to construct. Our results reveal that repetition dynamics, not scale alone, shape whether small-scale mixture experiments generalize. More broadly, they suggest that data repetition deserves treatment as a first-class variable in mixture optimization, rather than an inconvenient side effect of limited data.

Multi-agent LLM systems routinely generate multiple candidate responses that are aggregated by an LLM judge. To reduce the dominant prefill cost in such pipelines, recent work advocates KV cache reuse across partially shared contexts and reports substantial speedups for generation agents. In this work, we show that these efficiency gains do not transfer uniformly to judge-centric inference. Across GSM8K, MMLU, and HumanEval, we find that reuse strategies that are effective for execution agents can severely perturb judge behavior: end-task accuracy may appear stable, yet the judge's selection becomes highly inconsistent with dense prefill. We quantify this risk using Judge Consistency Rate (JCR) and provide diagnostics showing that reuse systematically weakens cross-candidate attention, especially for later candidate blocks. Our ablation further demonstrates that explicit cross-candidate interaction is crucial for preserving dense-prefill decisions. Overall, our results identify a previously overlooked failure mode of KV cache reuse and highlight judge-centric inference as a distinct regime that demands dedicated, risk-aware system design.

Progress in genomic foundation models is difficult to assess due to fragmented benchmarks, incompatible evaluation protocols, and task-specific reporting. As a result, claims of superiority or generality across models are often not directly comparable. We introduce GENEB, a large-scale diagnostic benchmark that evaluates frozen representations from 40 genomic foundation models across 100 tasks spanning 13 functional categories under a unified probing-based protocol, including few-shot regimes. GENEB enables controlled comparison across model scale, architecture, tokenization, and pretraining data while explicitly exposing task-level trade-offs. Our analysis shows that aggregate leaderboards are unstable: model rankings vary sharply across task categories, scale provides only modest and inconsistent gains, and architectural and pretraining alignment frequently outweigh parameter count. These results highlight limitations of current evaluation practices and position GENEB as a reference framework for principled comparison and category-aware model selection in genomic machine learning.

In Machine Translation (MT) evaluation, metric performance is assessed based on agreement with human judgments. In recent years, automatic metrics have demonstrated increasingly high levels of agreement with humans. To gain a clearer understanding of metric performance and establish an upper bound, we incorporate human baselines in the MT meta-evaluation, that is, the assessment of MT metrics' capabilities. Our results show that human annotators are not consistently superior to automatic metrics, with state-of-the-art metrics often ranking on par with or higher than human baselines. Despite these findings suggesting human parity, we discuss several reasons for caution. Finally, we explore the broader implications of our results for the research field, asking: Can we still reliably measure improvements in MT evaluation? With this work, we aim to shed light on the limits of our ability to measure progress in the field, fostering discussion on an issue that we believe is crucial to the entire MT evaluation community.

Language models are increasingly used in scientific discovery to generate hypotheses, propose candidate solutions, implement systems, and iteratively refine them. At the core of these trial-and-error loops lies evaluation: the process of obtaining feedback on candidate solutions via verifiers, simulators, or task-specific scoring functions. While prior work has highlighted the importance of evaluation, it has not explicitly formulated the problem of how evaluation-driven discovery loops can be scaled up in a principled and effective manner to push the boundaries of scientific discovery, a problem this paper seeks to address. We introduce Simple Test-time Evaluation-driven Scaling (SimpleTES), a general framework that strategically combines parallel exploration, feedback-driven refinement, and local selection, revealing substantial gains unlocked by scaling evaluation-driven discovery loops along the right dimensions. Across 21 scientific problems spanning six domains, SimpleTES discovers state-of-the-art solutions using gpt-oss models, consistently outperforming both frontier-model baselines and sophisticated optimization pipelines. Particularly, we sped up the widely used LASSO algorithm by over 2x, designed quantum circuit routing policies that reduce gate overhead by 24.5%, and discovered new Erdos minimum overlap constructions that surpass the best-known results. Beyond novel discoveries, SimpleTES produces trajectory-level histories that naturally supervise feedback-driven learning. When post-trained on successful trajectories, models not only improve efficiency on seen problems but also generalize to unseen problems, discovering solutions that base models fail to uncover. Together, our results establish effective evaluation-driven loop scaling as a central axis for advancing LLM-driven scientific discovery, and provide a simple yet practical framework for realizing these gains.

Patient recruitment is challenging for clinical trials. We introduce TrialGPT, an end-to-end framework for zero-shot patient-to-trial matching with large language models. TrialGPT comprises three modules: it first performs large-scale filtering to retrieve candidate trials (TrialGPT-Retrieval); then predicts criterion-level patient eligibility (TrialGPT-Matching); and finally generates trial-level scores (TrialGPT-Ranking). We evaluate TrialGPT on three cohorts of 183 synthetic patients with over 75,000 trial annotations. TrialGPT-Retrieval can recall over 90% of relevant trials using less than 6% of the initial collection. Manual evaluations on 1,015 patient-criterion pairs show that TrialGPT-Matching achieves an accuracy of 87.3% with faithful explanations, close to the expert performance. The TrialGPT-Ranking scores are highly correlated with human judgments and outperform the best-competing models by 43.8% in ranking and excluding trials. Furthermore, our user study reveals that TrialGPT can reduce the screening time by 42.6% in patient recruitment. Overall, these results have demonstrated promising opportunities for patient-to-trial matching with TrialGPT.

Test-time compute scaling is a primary axis for improving LLM reasoning. Existing methods primarily scale depth by extending a single reasoning trace. Scaling breadth by sampling multiple candidates in parallel is straightforward, but introduces a selection bottleneck: choosing the best candidate without a ground-truth verifier, since pointwise LLM judging is noisy and biased. To address this, we introduce OpenDeepThink, a population-based test-time compute framework that selects via pairwise Bradley-Terry comparison. Each generation, the LLM judges random pairs of candidates and aggregates votes via Bradley-Terry into a global ranking; top-ranked candidates are preserved and the top three quarters are mutated using the natural-language critiques produced during comparison; the bottom quarter is discarded. OpenDeepThink raises Gemini 3.1 Pro's effective Codeforces Elo by +405 points in eight sequential LLM-call rounds (~27 minutes wall-clock). The pipeline transfers across weaker and stronger models without retuning, and on the multi-domain HLE benchmark, gains appear concentrated in objectively verifiable domains and reverse in subjective ones. We release CF-73, a curated set of 73 expert-rated Codeforces problems with International Grandmaster annotation and 99% local-evaluation agreement against the official verdict.

Automatic metrics are now central to evaluating text-to-image models, often substituting for human judgment in benchmarking and large-scale filtering. However, it remains unclear whether these metrics truly prioritize semantic correctness or instead favor visually and socially prototypical images learned from biased data distributions. We identify and study prototypicality bias as a systematic failure mode in multimodal evaluation. We introduce a controlled contrastive benchmark ProtoBias (Prototypical Bias), spanning Animals, Objects, and Demography images, where semantically correct but non-prototypical images are paired with subtly incorrect yet prototypical adversarial counterparts. This setup enables a directional evaluation of whether metrics follow textual semantics or default to prototypes. Our results show that widely used metrics, including CLIPScore, PickScore, and VQA-based scores, frequently misrank these pairs, while even LLM-as-Judge systems exhibit uneven robustness in socially grounded cases. Human evaluations consistently favour semantic correctness with larger decision margins. Motivated by these findings, we propose ProtoScore, a robust 7B-parameter metric that substantially reduces failure rates and suppresses misranking, while running at orders of magnitude faster than the inference time of GPT-5, approaching the robustness of much larger closed-source judges.