arxiv_id float64 1.5k 2.51k | title stringlengths 9 178 ⌀ | authors stringlengths 2 22.8k | categories stringlengths 4 146 | summary stringlengths 103 1.92k ⌀ | published stringdate 2015-02-06 10:44:00 2025-07-10 17:59:58 ⌀ | comments stringlengths 2 417 ⌀ | journal_ref stringclasses 321
values | doi stringclasses 398
values | ss_title stringlengths 8 159 ⌀ | ss_authors stringlengths 11 8.38k ⌀ | ss_year float64 2.02k 2.03k ⌀ | ss_venue stringclasses 281
values | ss_citationCount float64 0 134k ⌀ | ss_referenceCount float64 0 429 ⌀ | ss_fieldsOfStudy stringclasses 47
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2,504.036 | MedSAM2: Segment Anything in 3D Medical Images and Videos | ['Jun Ma', 'Zongxin Yang', 'Sumin Kim', 'Bihui Chen', 'Mohammed Baharoon', 'Adibvafa Fallahpour', 'Reza Asakereh', 'Hongwei Lyu', 'Bo Wang'] | ['eess.IV', 'cs.AI', 'cs.CV'] | Medical image and video segmentation is a critical task for precision
medicine, which has witnessed considerable progress in developing task or
modality-specific and generalist models for 2D images. However, there have been
limited studies on building general-purpose models for 3D images and videos
with comprehensive u... | 2025-04-04T17:13:37Z | https://medsam2.github.io/ | null | null | MedSAM2: Segment Anything in 3D Medical Images and Videos | ['Jun Ma', 'Zongxin Yang', 'Sumin Kim', 'Bihui Chen', 'Mohammed Baharoon', 'Adibvafa Fallahpour', 'Reza Asakereh', 'Hongwei Lyu', 'Bo Wang'] | 2,025 | arXiv.org | 0 | 53 | ['Computer Science', 'Engineering'] |
2,504.03601 | APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated
Agent-Human Interplay | ['Akshara Prabhakar', 'Zuxin Liu', 'Ming Zhu', 'Jianguo Zhang', 'Tulika Awalgaonkar', 'Shiyu Wang', 'Zhiwei Liu', 'Haolin Chen', 'Thai Hoang', 'Juan Carlos Niebles', 'Shelby Heinecke', 'Weiran Yao', 'Huan Wang', 'Silvio Savarese', 'Caiming Xiong'] | ['cs.CL', 'cs.AI', 'cs.LG'] | Training effective AI agents for multi-turn interactions requires
high-quality data that captures realistic human-agent dynamics, yet such data
is scarce and expensive to collect manually. We introduce APIGen-MT, a
two-phase framework that generates verifiable and diverse multi-turn agent
data. In the first phase, our ... | 2025-04-04T17:13:57Z | 12 pages plus references and appendices | null | null | APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplay | ['Akshara Prabhakar', 'Zuxin Liu', 'Weiran Yao', 'Jianguo Zhang', 'Ming Zhu', 'Shiyu Wang', 'Zhiwei Liu', 'T. Awalgaonkar', 'Haolin Chen', 'Thai Hoang', 'Juan Carlos Niebles', 'Shelby Heinecke', 'Huan Wang', 'Silvio Savarese', 'Caiming Xiong'] | 2,025 | arXiv.org | 11 | 52 | ['Computer Science'] |
2,504.03624 | Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer
Models | ['NVIDIA', ':', 'Aaron Blakeman', 'Aarti Basant', 'Abhinav Khattar', 'Adithya Renduchintala', 'Akhiad Bercovich', 'Aleksander Ficek', 'Alexis Bjorlin', 'Ali Taghibakhshi', 'Amala Sanjay Deshmukh', 'Ameya Sunil Mahabaleshwarkar', 'Andrew Tao', 'Anna Shors', 'Ashwath Aithal', 'Ashwin Poojary', 'Ayush Dattagupta', 'Balara... | ['cs.CL', 'cs.AI', 'cs.LG'] | As inference-time scaling becomes critical for enhanced reasoning
capabilities, it is increasingly becoming important to build models that are
efficient to infer. We introduce Nemotron-H, a family of 8B and 56B/47B hybrid
Mamba-Transformer models designed to reduce inference cost for a given accuracy
level. To achieve ... | 2025-04-04T17:41:58Z | null | null | null | Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models | ['Nvidia Aaron Blakeman', 'Aarti Basant', 'Abhinav Khattar', 'Adi Renduchintala', 'A. Bercovich', 'Aleksander Ficek', 'Alexis Bjorlin', 'Ali Taghibakhshi', 'Amala Sanjay Deshmukh', 'A. Mahabaleshwarkar', 'Andrew Tao', 'Anna C. Shors', 'Ashwath Aithal', 'Ashwin Poojary', 'Ayush Dattagupta', 'B. Buddharaju', 'Bobby Chen'... | 2,025 | arXiv.org | 8 | 0 | ['Computer Science'] |
2,504.03709 | Ocularone-Bench: Benchmarking DNN Models on GPUs to Assist the Visually
Impaired | ['Suman Raj', 'Bhavani A Madhabhavi', 'Kautuk Astu', 'Arnav A Rajesh', 'Pratham M', 'Yogesh Simmhan'] | ['cs.DC'] | VIP navigation requires multiple DNN models for identification, posture
analysis, and depth estimation to ensure safe mobility. Using a hazard vest as
a unique identifier enhances visibility while selecting the right DNN model and
computing device balances accuracy and real-time performance. We present
Ocularone-Bench,... | 2025-03-27T10:08:18Z | 11 pages, 6 figures, To Appear at the IEEE Workshop on Parallel and
Distributed Processing for Computational Social Systems (ParSocial),
Co-located with IEEE IPDPS 2025 | null | null | null | null | null | null | null | null | null |
2,504.03964 | Clinical ModernBERT: An efficient and long context encoder for
biomedical text | ['Simon A. Lee', 'Anthony Wu', 'Jeffrey N. Chiang'] | ['cs.CL', 'cs.AI', 'cs.LG'] | We introduce Clinical ModernBERT, a transformer based encoder pretrained on
large scale biomedical literature, clinical notes, and medical ontologies,
incorporating PubMed abstracts, MIMIC IV clinical data, and medical codes with
their textual descriptions. Building on ModernBERT the current state of the art
natural la... | 2025-04-04T22:14:12Z | Manuscript writeup corresponding to the Clinical ModernBERT
pre-trained encoder (https://huggingface.co/Simonlee711/Clinical_ModernBERT) | null | null | null | null | null | null | null | null | null |
2,504.0406 | VocalNet: Speech LLM with Multi-Token Prediction for Faster and
High-Quality Generation | ['Yuhao Wang', 'Heyang Liu', 'Ziyang Cheng', 'Ronghua Wu', 'Qunshan Gu', 'Yanfeng Wang', 'Yu Wang'] | ['cs.CL', 'cs.AI', 'cs.SD', 'eess.AS'] | Speech large language models (LLMs) have emerged as a prominent research
focus in speech processing. We introduce VocalNet-1B and VocalNet-8B, a series
of high-performance, low-latency speech LLMs enabled by a scalable and
model-agnostic training framework designed for real-time voice interaction.
Central to our contri... | 2025-04-05T04:57:12Z | null | null | null | null | null | null | null | null | null | null |
2,504.04131 | Precise Legal Sentence Boundary Detection for Retrieval at Scale:
NUPunkt and CharBoundary | ['Michael J Bommarito', 'Daniel Martin Katz', 'Jillian Bommarito'] | ['cs.CL'] | We present NUPunkt and CharBoundary, two sentence boundary detection
libraries optimized for high-precision, high-throughput processing of legal
text in large-scale applications such as due diligence, e-discovery, and legal
research. These libraries address the critical challenges posed by legal
documents containing sp... | 2025-04-05T10:48:34Z | 12 pages, 5 figures, 6 tables | null | null | null | null | null | null | null | null | null |
2,504.04158 | JarvisIR: Elevating Autonomous Driving Perception with Intelligent Image
Restoration | ['Yunlong Lin', 'Zixu Lin', 'Haoyu Chen', 'Panwang Pan', 'Chenxin Li', 'Sixiang Chen', 'Yeying Jin', 'Wenbo Li', 'Xinghao Ding'] | ['cs.CV'] | Vision-centric perception systems struggle with unpredictable and coupled
weather degradations in the wild. Current solutions are often limited, as they
either depend on specific degradation priors or suffer from significant domain
gaps. To enable robust and autonomous operation in real-world conditions, we
propose Jar... | 2025-04-05T12:38:55Z | 25 pages, 15 figures | null | null | JarvisIR: Elevating Autonomous Driving Perception with Intelligent Image Restoration | ['Yunlong Lin', 'Zixu Lin', 'Haoyu Chen', 'Panwang Pan', 'Chenxin Li', 'Sixiang Chen', 'Yeying Jin', 'Wenbo Li', 'Xinghao Ding'] | 2,025 | arXiv.org | 2 | 102 | ['Computer Science'] |
2,504.04323 | MedM-VL: What Makes a Good Medical LVLM? | ['Yiming Shi', 'Shaoshuai Yang', 'Xun Zhu', 'Haoyu Wang', 'Miao Li', 'Ji Wu'] | ['cs.CV'] | Medical image analysis is essential in modern healthcare. Deep learning has
redirected research focus toward complex medical multimodal tasks, including
report generation and visual question answering. Traditional task-specific
models often fall short in handling these challenges. Large vision-language
models (LVLMs) o... | 2025-04-06T01:44:46Z | null | null | null | null | null | null | null | null | null | null |
2,504.04377 | PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages | ['Priyanshu Kumar', 'Devansh Jain', 'Akhila Yerukola', 'Liwei Jiang', 'Himanshu Beniwal', 'Thomas Hartvigsen', 'Maarten Sap'] | ['cs.CL'] | Truly multilingual safety moderation efforts for Large Language Models (LLMs)
have been hindered by a narrow focus on a small set of languages (e.g.,
English, Chinese) as well as a limited scope of safety definition, resulting in
significant gaps in moderation capabilities. To bridge these gaps, we release
POLYGUARD, a... | 2025-04-06T06:09:21Z | null | null | null | null | null | null | null | null | null | null |
2,504.04524 | Trust Region Preference Approximation: A simple and stable reinforcement
learning algorithm for LLM reasoning | ['Xuerui Su', 'Shufang Xie', 'Guoqing Liu', 'Yingce Xia', 'Renqian Luo', 'Peiran Jin', 'Zhiming Ma', 'Yue Wang', 'Zun Wang', 'Yuting Liu'] | ['cs.LG', 'cs.AI'] | Recently, Large Language Models (LLMs) have rapidly evolved, approaching
Artificial General Intelligence (AGI) while benefiting from large-scale
reinforcement learning to enhance Human Alignment (HA) and Reasoning. Recent
reward-based optimization algorithms, such as Proximal Policy Optimization
(PPO) and Group Relativ... | 2025-04-06T15:48:26Z | 10pages | null | null | null | null | null | null | null | null | null |
2,504.04704 | LagKV: Lag-Relative Information of the KV Cache Tells Which Tokens Are
Important | ['Manlai Liang', 'JiaMing Zhang', 'Xiong Li', 'Jinlong Li'] | ['cs.LG', 'cs.AI', 'cs.CL', 'cs.CV'] | The increasing size of the Key-Value (KV) cache during the Large Language
Models long-context inference is the main obstacle for its balance between the
deployment cost and task accuracy. To reduce the KV cache size in such
scenarios, most previous efforts leveraged on the attention weight to evict
non-critical cache t... | 2025-04-07T03:22:15Z | null | null | null | LagKV: Lag-Relative Information of the KV Cache Tells Which Tokens Are Important | ['Manlai Liang', 'JiaMing Zhang', 'Xiong Li', 'Jinlong Li'] | 2,025 | arXiv.org | 1 | 23 | ['Computer Science'] |
2,504.04823 | Quantization Hurts Reasoning? An Empirical Study on Quantized Reasoning
Models | ['Ruikang Liu', 'Yuxuan Sun', 'Manyi Zhang', 'Haoli Bai', 'Xianzhi Yu', 'Tiezheng Yu', 'Chun Yuan', 'Lu Hou'] | ['cs.CL', 'cs.AI'] | Recent advancements in reasoning language models have demonstrated remarkable
performance in complex tasks, but their extended chain-of-thought reasoning
process increases inference overhead. While quantization has been widely
adopted to reduce the inference cost of large language models, its impact on
reasoning models... | 2025-04-07T08:22:45Z | null | null | null | null | null | null | null | null | null | null |
2,504.04842 | FantasyTalking: Realistic Talking Portrait Generation via Coherent
Motion Synthesis | ['Mengchao Wang', 'Qiang Wang', 'Fan Jiang', 'Yaqi Fan', 'Yunpeng Zhang', 'Yonggang Qi', 'Kun Zhao', 'Mu Xu'] | ['cs.CV'] | Creating a realistic animatable avatar from a single static portrait remains
challenging. Existing approaches often struggle to capture subtle facial
expressions, the associated global body movements, and the dynamic background.
To address these limitations, we propose a novel framework that leverages a
pretrained vide... | 2025-04-07T08:56:01Z | null | null | null | null | null | null | null | null | null | null |
2,504.04907 | Video-Bench: Human-Aligned Video Generation Benchmark | ['Hui Han', 'Siyuan Li', 'Jiaqi Chen', 'Yiwen Yuan', 'Yuling Wu', 'Chak Tou Leong', 'Hanwen Du', 'Junchen Fu', 'Youhua Li', 'Jie Zhang', 'Chi Zhang', 'Li-jia Li', 'Yongxin Ni'] | ['cs.CV', 'cs.AI'] | Video generation assessment is essential for ensuring that generative models
produce visually realistic, high-quality videos while aligning with human
expectations. Current video generation benchmarks fall into two main
categories: traditional benchmarks, which use metrics and embeddings to
evaluate generated video qua... | 2025-04-07T10:32:42Z | Accepted by CVPR'25 | null | null | null | null | null | null | null | null | null |
2,504.04945 | A Llama walks into the 'Bar': Efficient Supervised Fine-Tuning for Legal
Reasoning in the Multi-state Bar Exam | ['Rean Fernandes', 'André Biedenkapp', 'Frank Hutter', 'Noor Awad'] | ['cs.LG', 'cs.AI', 'cs.CL', 'I.2.7; I.2.1'] | Legal reasoning tasks present unique challenges for large language models
(LLMs) due to the complexity of domain-specific knowledge and reasoning
processes. This paper investigates how effectively smaller language models
(Llama 2 7B and Llama 3 8B) can be fine-tuned with a limited dataset of 1,514
Multi-state Bar Exami... | 2025-04-07T11:31:22Z | COLM 2025 preprint, 9 pages, 3 figures, 16 appendix pages | null | null | null | null | null | null | null | null | null |
2,504.04949 | One Quantizer is Enough: Toward a Lightweight Audio Codec | ['Linwei Zhai', 'Han Ding', 'Cui Zhao', 'fei wang', 'Ge Wang', 'Wang Zhi', 'Wei Xi'] | ['cs.SD', 'cs.AI', '68T07', 'I.2.m'] | Neural audio codecs have recently gained traction for their ability to
compress high-fidelity audio and generate discrete tokens that can be utilized
in downstream generative modeling tasks. However, leading approaches often rely
on resource-intensive models and multi-quantizer architectures, resulting in
considerable ... | 2025-04-07T11:34:39Z | null | null | null | One Quantizer is Enough: Toward a Lightweight Audio Codec | ['Linwei Zhai', 'H. Ding', 'Cui Zhao', 'Fei Wang', 'Ge Wang', 'Wang Zhi', 'Wei Xi'] | 2,025 | arXiv.org | 0 | 36 | ['Computer Science'] |
2,504.04953 | M-Prometheus: A Suite of Open Multilingual LLM Judges | ['José Pombal', 'Dongkeun Yoon', 'Patrick Fernandes', 'Ian Wu', 'Seungone Kim', 'Ricardo Rei', 'Graham Neubig', 'André F. T. Martins'] | ['cs.CL', 'cs.AI'] | The use of language models for automatically evaluating long-form text
(LLM-as-a-judge) is becoming increasingly common, yet most LLM judges are
optimized exclusively for English, with strategies for enhancing their
multilingual evaluation capabilities remaining largely unexplored in the
current literature. This has cr... | 2025-04-07T11:37:26Z | null | null | null | M-Prometheus: A Suite of Open Multilingual LLM Judges | ['José P. Pombal', 'Dongkeun Yoon', 'Patrick Fernandes', 'Ian Wu', 'Seungone Kim', 'Ricardo Rei', 'Graham Neubig', "Andr'e F. T. Martins"] | 2,025 | arXiv.org | 5 | 0 | ['Computer Science'] |
2,504.05089 | Climplicit: Climatic Implicit Embeddings for Global Ecological Tasks | ['Johannes Dollinger', 'Damien Robert', 'Elena Plekhanova', 'Lukas Drees', 'Jan Dirk Wegner'] | ['cs.CV'] | Deep learning on climatic data holds potential for macroecological
applications. However, its adoption remains limited among scientists outside
the deep learning community due to storage, compute, and technical expertise
barriers. To address this, we introduce Climplicit, a spatio-temporal
geolocation encoder pretraine... | 2025-04-07T13:58:55Z | Published as a workshop paper at "Tackling Climate Change with
Machine Learning", ICLR 2025 | null | null | null | null | null | null | null | null | null |
2,504.0515 | A Reinforcement Learning Method for Environments with Stochastic
Variables: Post-Decision Proximal Policy Optimization with Dual Critic
Networks | ['Leonardo Kanashiro Felizardo', 'Edoardo Fadda', 'Paolo Brandimarte', 'Emilio Del-Moral-Hernandez', 'Mariá Cristina Vasconcelos Nascimento'] | ['cs.LG', 'cs.AI', 'I.2.6; G.1.6'] | This paper presents Post-Decision Proximal Policy Optimization (PDPPO), a
novel variation of the leading deep reinforcement learning method, Proximal
Policy Optimization (PPO). The PDPPO state transition process is divided into
two steps: a deterministic step resulting in the post-decision state and a
stochastic step l... | 2025-04-07T14:56:43Z | 12 pages, 4 figures. Accepted for presentation at IJCNN 2025 | null | null | null | null | null | null | null | null | null |
2,504.05186 | Training state-of-the-art pathology foundation models with orders of
magnitude less data | ['Mikhail Karasikov', 'Joost van Doorn', 'Nicolas Känzig', 'Melis Erdal Cesur', 'Hugo Mark Horlings', 'Robert Berke', 'Fei Tang', 'Sebastian Otálora'] | ['cs.CV', 'cs.LG'] | The field of computational pathology has recently seen rapid advances driven
by the development of modern vision foundation models (FMs), typically trained
on vast collections of pathology images. Recent studies demonstrate that
increasing the training data set and model size and integrating domain-specific
image proce... | 2025-04-07T15:38:12Z | 10 pages, 3 figures | null | null | null | null | null | null | null | null | null |
2,504.05299 | SmolVLM: Redefining small and efficient multimodal models | ['Andrés Marafioti', 'Orr Zohar', 'Miquel Farré', 'Merve Noyan', 'Elie Bakouch', 'Pedro Cuenca', 'Cyril Zakka', 'Loubna Ben Allal', 'Anton Lozhkov', 'Nouamane Tazi', 'Vaibhav Srivastav', 'Joshua Lochner', 'Hugo Larcher', 'Mathieu Morlon', 'Lewis Tunstall', 'Leandro von Werra', 'Thomas Wolf'] | ['cs.AI', 'cs.CV'] | Large Vision-Language Models (VLMs) deliver exceptional performance but
require significant computational resources, limiting their deployment on
mobile and edge devices. Smaller VLMs typically mirror design choices of larger
models, such as extensive image tokenization, leading to inefficient GPU memory
usage and cons... | 2025-04-07T17:58:57Z | null | null | null | SmolVLM: Redefining small and efficient multimodal models | ['Andrés Marafioti', 'Orr Zohar', "Miquel Farr'e", 'Merve Noyan', 'Elie Bakouch', 'Pedro Cuenca', 'Cyril Zakka', 'Loubna Ben Allal', 'Anton Lozhkov', 'Nouamane Tazi', 'Vaibhav Srivastav', 'Joshua Lochner', 'Hugo Larcher', 'Mathieu Morlon', 'Lewis Tunstall', 'L. V. Werra', 'Thomas Wolf'] | 2,025 | arXiv.org | 16 | 83 | ['Computer Science'] |
2,504.05304 | Gaussian Mixture Flow Matching Models | ['Hansheng Chen', 'Kai Zhang', 'Hao Tan', 'Zexiang Xu', 'Fujun Luan', 'Leonidas Guibas', 'Gordon Wetzstein', 'Sai Bi'] | ['cs.LG', 'cs.CV'] | Diffusion models approximate the denoising distribution as a Gaussian and
predict its mean, whereas flow matching models reparameterize the Gaussian mean
as flow velocity. However, they underperform in few-step sampling due to
discretization error and tend to produce over-saturated colors under
classifier-free guidance... | 2025-04-07T17:59:42Z | ICML 2025. Code: https://github.com/Lakonik/GMFlow | null | null | null | null | null | null | null | null | null |
2,504.05402 | Time-adaptive Video Frame Interpolation based on Residual Diffusion | ['Victor Fonte Chavez', 'Claudia Esteves', 'Jean-Bernard Hayet'] | ['cs.CV'] | In this work, we propose a new diffusion-based method for video frame
interpolation (VFI), in the context of traditional hand-made animation. We
introduce three main contributions: The first is that we explicitly handle the
interpolation time in our model, which we also re-estimate during the training
process, to cope ... | 2025-04-07T18:15:45Z | 17 pages | null | null | Time-adaptive Video Frame Interpolation based on Residual Diffusion | ['Victor Fonte Chavez', 'Claudia Esteves', 'J. Hayet'] | 2,025 | arXiv.org | 0 | 32 | ['Computer Science'] |
2,504.05523 | Pretraining Language Models for Diachronic Linguistic Change Discovery | ['Elisabeth Fittschen', 'Sabrina Li', 'Tom Lippincott', 'Leshem Choshen', 'Craig Messner'] | ['cs.CL'] | Large language models (LLMs) have shown potential as tools for scientific
discovery. This has engendered growing interest in their use in humanistic
disciplines, such as historical linguistics and literary studies. These fields
often construct arguments on the basis of delineations like genre, or more
inflexibly, time ... | 2025-04-07T21:51:32Z | null | null | null | null | null | null | null | null | null | null |
2,504.05535 | COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for
Alignment with Human Values | ['M-A-P Team', 'Siwei Wu', 'Jincheng Ren', 'Xinrun Du', 'Shuyue Guo', 'Xingwei Qu', 'Yiming Liang', 'Jie Liu', 'Yunwen Li', 'Tianyu Zheng', 'Boyu Feng', 'Huaqing Yuan', 'Zenith Wang', 'Jiaheng Liu', 'Wenhao Huang', 'Chenglin Cai', 'Haoran Que', 'Jian Yang', 'Yuelin Bai', 'Zekun Moore Wang', 'Zhouliang Yu', 'Qunshu Lin'... | ['cs.CL'] | Aligning large language models (LLMs) with human preferences has achieved
remarkable success. However, existing Chinese preference datasets are limited
by small scale, narrow domain coverage, and lack of rigorous data validation.
Additionally, the reliance on human annotators for instruction and response
labeling signi... | 2025-04-07T22:15:51Z | null | null | null | null | null | null | null | null | null | null |
2,504.05579 | TAPNext: Tracking Any Point (TAP) as Next Token Prediction | ['Artem Zholus', 'Carl Doersch', 'Yi Yang', 'Skanda Koppula', 'Viorica Patraucean', 'Xu Owen He', 'Ignacio Rocco', 'Mehdi S. M. Sajjadi', 'Sarath Chandar', 'Ross Goroshin'] | ['cs.CV'] | Tracking Any Point (TAP) in a video is a challenging computer vision problem
with many demonstrated applications in robotics, video editing, and 3D
reconstruction. Existing methods for TAP rely heavily on complex
tracking-specific inductive biases and heuristics, limiting their generality
and potential for scaling. To ... | 2025-04-08T00:28:42Z | null | null | null | null | null | null | null | null | null | null |
2,504.05599 | Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thought | ['Yi Peng', 'Peiyu Wang', 'Xiaokun Wang', 'Yichen Wei', 'Jiangbo Pei', 'Weijie Qiu', 'Ai Jian', 'Yunzhuo Hao', 'Jiachun Pan', 'Tianyidan Xie', 'Li Ge', 'Rongxian Zhuang', 'Xuchen Song', 'Yang Liu', 'Yahui Zhou'] | ['cs.CV', 'cs.CL'] | We introduce Skywork R1V, a multimodal reasoning model extending the an
R1-series Large language models (LLM) to visual modalities via an efficient
multimodal transfer method. Leveraging a lightweight visual projector, Skywork
R1V facilitates seamless multimodal adaptation without necessitating retraining
of either the... | 2025-04-08T01:19:20Z | null | null | null | null | null | null | null | null | null | null |
2,504.05741 | DDT: Decoupled Diffusion Transformer | ['Shuai Wang', 'Zhi Tian', 'Weilin Huang', 'Limin Wang'] | ['cs.CV', 'cs.AI'] | Diffusion transformers have demonstrated remarkable generation quality,
albeit requiring longer training iterations and numerous inference steps. In
each denoising step, diffusion transformers encode the noisy inputs to extract
the lower-frequency semantic component and then decode the higher frequency
with identical m... | 2025-04-08T07:17:45Z | sota on ImageNet256 and ImageNet512 | null | null | null | null | null | null | null | null | null |
2,504.05747 | SEA-LION: Southeast Asian Languages in One Network | ['Raymond Ng', 'Thanh Ngan Nguyen', 'Yuli Huang', 'Ngee Chia Tai', 'Wai Yi Leong', 'Wei Qi Leong', 'Xianbin Yong', 'Jian Gang Ngui', 'Yosephine Susanto', 'Nicholas Cheng', 'Hamsawardhini Rengarajan', 'Peerat Limkonchotiwat', 'Adithya Venkatadri Hulagadri', 'Kok Wai Teng', 'Yeo Yeow Tong', 'Bryan Siow', 'Wei Yi Teo', 'W... | ['cs.CL'] | Recently, Large Language Models (LLMs) have dominated much of the artificial
intelligence scene with their ability to process and generate natural
languages. However, the majority of LLM research and development remains
English-centric, leaving low-resource languages such as those in the Southeast
Asian (SEA) region un... | 2025-04-08T07:24:51Z | We released our model at
https://huggingface.co/collections/aisingapore/sea-lionv3-672589a39cdadd6a5b199581 | null | null | null | null | null | null | null | null | null |
2,504.06196 | TxGemma: Efficient and Agentic LLMs for Therapeutics | ['Eric Wang', 'Samuel Schmidgall', 'Paul F. Jaeger', 'Fan Zhang', 'Rory Pilgrim', 'Yossi Matias', 'Joelle Barral', 'David Fleet', 'Shekoofeh Azizi'] | ['cs.AI', 'cs.CL', 'cs.LG'] | Therapeutic development is a costly and high-risk endeavor that is often
plagued by high failure rates. To address this, we introduce TxGemma, a suite
of efficient, generalist large language models (LLMs) capable of therapeutic
property prediction as well as interactive reasoning and explainability. Unlike
task-specifi... | 2025-04-08T16:39:02Z | null | null | null | TxGemma: Efficient and Agentic LLMs for Therapeutics | ['Eric Wang', 'Samuel Schmidgall', 'Paul F. Jaeger', 'Fan Zhang', 'Rory Pilgrim', 'Yossi Matias', 'Joelle Barral', 'David Fleet', 'Shekoofeh Azizi'] | 2,025 | arXiv.org | 13 | 0 | ['Computer Science'] |
2,504.06214 | From 128K to 4M: Efficient Training of Ultra-Long Context Large Language
Models | ['Chejian Xu', 'Wei Ping', 'Peng Xu', 'Zihan Liu', 'Boxin Wang', 'Mohammad Shoeybi', 'Bo Li', 'Bryan Catanzaro'] | ['cs.CL', 'cs.AI', 'cs.LG'] | Long-context capabilities are essential for a wide range of applications,
including document and video understanding, in-context learning, and
inference-time scaling, all of which require models to process and reason over
long sequences of text and multimodal data. In this work, we introduce a
efficient training recipe... | 2025-04-08T16:58:58Z | null | null | null | From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models | ['Chejian Xu', 'Wei Ping', 'Peng Xu', 'Zihan Liu', 'Boxin Wang', 'M. Shoeybi', 'Bo Li', 'Bryan Catanzaro'] | 2,025 | arXiv.org | 4 | 41 | ['Computer Science'] |
2,504.0622 | Earth-Adapter: Bridge the Geospatial Domain Gaps with Mixture of
Frequency Adaptation | ['Xiaoxing Hu', 'Ziyang Gong', 'Yupei Wang', 'Yuru Jia', 'Gen Luo', 'Xue Yang'] | ['cs.CV'] | Parameter-Efficient Fine-Tuning (PEFT) is a technique that allows us to adapt
powerful Foundation Models (FMs) to diverse downstream tasks while preserving
and unleashing their inherent capabilities. However, we have observed that
existing PEFT methods, which are often designed with natural imagery in mind,
struggle wh... | 2025-04-08T17:09:33Z | null | null | null | null | null | null | null | null | null | null |
2,504.06225 | Encoder-Decoder Gemma: Improving the Quality-Efficiency Trade-Off via
Adaptation | ['Biao Zhang', 'Fedor Moiseev', 'Joshua Ainslie', 'Paul Suganthan', 'Min Ma', 'Surya Bhupatiraju', 'Fede Lebron', 'Orhan Firat', 'Armand Joulin', 'Zhe Dong'] | ['cs.CL', 'cs.LG'] | While decoder-only large language models (LLMs) have shown impressive
results, encoder-decoder models are still widely adopted in real-world
applications for their inference efficiency and richer encoder representation.
In this paper, we study a novel problem: adapting pretrained decoder-only LLMs
to encoder-decoder, w... | 2025-04-08T17:13:41Z | null | null | null | Encoder-Decoder Gemma: Improving the Quality-Efficiency Trade-Off via Adaptation | ['Biao Zhang', 'Fedor Moiseev', 'Joshua Ainslie', 'P. Suganthan', 'Min Ma', 'Surya Bhupatiraju', 'Federico Lebron', 'Orhan Firat', 'Armand Joulin', 'Zhe Dong'] | 2,025 | arXiv.org | 0 | 54 | ['Computer Science'] |
2,504.0633 | Analyzing the Impact of Low-Rank Adaptation for Cross-Domain Few-Shot
Object Detection in Aerial Images | ['Hicham Talaoubrid', 'Anissa Mokraoui', 'Ismail Ben Ayed', 'Axel Prouvost', 'Sonimith Hang', 'Monit Korn', 'Rémi Harvey'] | ['cs.CV', 'cs.AI'] | This paper investigates the application of Low-Rank Adaptation (LoRA) to
small models for cross-domain few-shot object detection in aerial images.
Originally designed for large-scale models, LoRA helps mitigate overfitting,
making it a promising approach for resource-constrained settings. We integrate
LoRA into Diffusi... | 2025-04-08T14:10:39Z | null | null | null | Analyzing the Impact of Low-Rank Adaptation for Cross-Domain Few-Shot Object Detection in Aerial Images | ['Hicham Talaoubrid', 'Anissa Zergaïnoh-Mokraoui', 'Ismail Ben Ayed', 'Axel Prouvost', 'Sonimith Hang', 'Monit Korn', "R'emi Harvey"] | 2,025 | arXiv.org | 1 | 25 | ['Computer Science'] |
2,504.06632 | PosterMaker: Towards High-Quality Product Poster Generation with
Accurate Text Rendering | ['Yifan Gao', 'Zihang Lin', 'Chuanbin Liu', 'Min Zhou', 'Tiezheng Ge', 'Bo Zheng', 'Hongtao Xie'] | ['cs.CV'] | Product posters, which integrate subject, scene, and text, are crucial
promotional tools for attracting customers. Creating such posters using modern
image generation methods is valuable, while the main challenge lies in
accurately rendering text, especially for complex writing systems like Chinese,
which contains over... | 2025-04-09T07:13:08Z | Accepted by CVPR 2025. Project Page: https://poster-maker.github.io | null | null | null | null | null | null | null | null | null |
2,504.06778 | CAFA: a Controllable Automatic Foley Artist | ['Roi Benita', 'Michael Finkelson', 'Tavi Halperin', 'Gleb Sterkin', 'Yossi Adi'] | ['cs.SD', 'eess.AS'] | Foley is a key element in video production, refers to the process of adding
an audio signal to a silent video while ensuring semantic and temporal
alignment. In recent years, the rise of personalized content creation and
advancements in automatic video-to-audio models have increased the demand for
greater user control ... | 2025-04-09T10:58:54Z | Renamed paper to "CAFA: a Controllable Automatic Foley Artist" from
"Controllable Automatic Foley Artist". Updated link to demo page | null | null | null | null | null | null | null | null | null |
2,504.06868 | Persona Dynamics: Unveiling the Impact of Personality Traits on Agents
in Text-Based Games | ['Seungwon Lim', 'Seungbeen Lee', 'Dongjun Min', 'Youngjae Yu'] | ['cs.CL', 'cs.AI'] | Artificial agents are increasingly central to complex interactions and
decision-making tasks, yet aligning their behaviors with desired human values
remains an open challenge. In this work, we investigate how human-like
personality traits influence agent behavior and performance within text-based
interactive environmen... | 2025-04-09T13:17:00Z | null | null | null | Persona Dynamics: Unveiling the Impact of Personality Traits on Agents in Text-Based Games | ['Seungwon Lim', 'Seungbeen Lee', 'Dongjun Min', 'Youngjae Yu'] | 2,025 | arXiv.org | 0 | 37 | ['Computer Science'] |
2,504.06895 | ColorizeDiffusion v2: Enhancing Reference-based Sketch Colorization
Through Separating Utilities | ['Dingkun Yan', 'Xinrui Wang', 'Yusuke Iwasawa', 'Yutaka Matsuo', 'Suguru Saito', 'Jiaxian Guo'] | ['cs.CV'] | Reference-based sketch colorization methods have garnered significant
attention due to their potential applications in the animation production
industry. However, most existing methods are trained with image triplets of
sketch, reference, and ground truth that are semantically and spatially
well-aligned, while real-wor... | 2025-04-09T13:55:32Z | null | null | null | null | null | null | null | null | null | null |
2,504.06958 | VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement
Fine-Tuning | ['Xinhao Li', 'Ziang Yan', 'Desen Meng', 'Lu Dong', 'Xiangyu Zeng', 'Yinan He', 'Yali Wang', 'Yu Qiao', 'Yi Wang', 'Limin Wang'] | ['cs.CV'] | Recent advancements in reinforcement learning have significantly advanced the
reasoning capabilities of multimodal large language models (MLLMs). While
approaches such as Group Relative Policy Optimization (GRPO) and rule-based
reward mechanisms demonstrate promise in text and image domains, their
application to video ... | 2025-04-09T15:09:27Z | null | null | null | VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning | ['Xinhao Li', 'Ziang Yan', 'Desen Meng', 'Lu Dong', 'Xiangyun Zeng', 'Yinan He', 'Yali Wang', 'Yu Qiao', 'Yi Wang', 'Limin Wang'] | 2,025 | arXiv.org | 38 | 45 | ['Computer Science'] |
2,504.06962 | Efficient Self-Supervised Learning for Earth Observation via Dynamic
Dataset Curation | ['Thomas Kerdreux', 'Alexandre Tuel', 'Quentin Febvre', 'Alexis Mouche', 'Bertrand Chapron'] | ['cs.CV', 'cs.AI'] | Self-supervised learning (SSL) has enabled the development of vision
foundation models for Earth Observation (EO), demonstrating strong
transferability across diverse remote sensing tasks. While prior work has
focused on network architectures and training strategies, the role of dataset
curation, especially in balancin... | 2025-04-09T15:13:26Z | Accepted at CVPR Workshop : The First Workshop on Foundation and
Large Vision Models in Remote Sensing | null | null | null | null | null | null | null | null | null |
2,504.07053 | TASTE: Text-Aligned Speech Tokenization and Embedding for Spoken
Language Modeling | ['Liang-Hsuan Tseng', 'Yi-Chang Chen', 'Kuan-Yi Lee', 'Da-Shan Shiu', 'Hung-yi Lee'] | ['cs.CL', 'cs.SD', 'eess.AS'] | Recent efforts target spoken language models (SLMs) that not only listen but
also speak for more natural human-LLM interaction. Joint speech-text modeling
is a promising direction to achieve this. However, the effectiveness of recent
speech tokens for joint modeling remains underexplored. To address this, we
introduce ... | 2025-04-09T17:14:33Z | Preprint | null | null | TASTE: Text-Aligned Speech Tokenization and Embedding for Spoken Language Modeling | ['Liang-Hsuan Tseng', 'Yi-Chang Chen', 'Kuan-Yi Lee', 'Da-shan Shiu', 'Hung-yi Lee'] | 2,025 | arXiv.org | 0 | 60 | ['Computer Science', 'Engineering'] |
2,504.07069 | HalluciNot: Hallucination Detection Through Context and Common Knowledge
Verification | ['Bibek Paudel', 'Alexander Lyzhov', 'Preetam Joshi', 'Puneet Anand'] | ['cs.CL', 'cs.AI'] | This paper introduces a comprehensive system for detecting hallucinations in
large language model (LLM) outputs in enterprise settings. We present a novel
taxonomy of LLM responses specific to hallucination in enterprise applications,
categorizing them into context-based, common knowledge, enterprise-specific,
and inno... | 2025-04-09T17:39:41Z | null | null | null | HalluciNot: Hallucination Detection Through Context and Common Knowledge Verification | ['B. Paudel', 'Alexander Lyzhov', 'Preetam Joshi', 'Puneet Anand'] | 2,025 | arXiv.org | 2 | 36 | ['Computer Science'] |
2,504.07089 | OmniCaptioner: One Captioner to Rule Them All | ['Yiting Lu', 'Jiakang Yuan', 'Zhen Li', 'Shitian Zhao', 'Qi Qin', 'Xinyue Li', 'Le Zhuo', 'Licheng Wen', 'Dongyang Liu', 'Yuewen Cao', 'Xiangchao Yan', 'Xin Li', 'Tianshuo Peng', 'Shufei Zhang', 'Botian Shi', 'Tao Chen', 'Zhibo Chen', 'Lei Bai', 'Peng Gao', 'Bo Zhang'] | ['cs.CV', 'cs.CL'] | We propose OmniCaptioner, a versatile visual captioning framework for
generating fine-grained textual descriptions across a wide variety of visual
domains. Unlike prior methods limited to specific image types (e.g., natural
images or geometric visuals), our framework provides a unified solution for
captioning natural i... | 2025-04-09T17:58:58Z | More visualizations on Homepage:
https://alpha-innovator.github.io/OmniCaptioner-project-page and Official
code: https://github.com/Alpha-Innovator/OmniCaptioner | null | null | OmniCaptioner: One Captioner to Rule Them All | ['Yiting Lu', 'Jiakang Yuan', 'Zhen Li', 'Shitian Zhao', 'Qi Qin', 'Xinyue Li', 'Le Zhuo', 'Licheng Wen', 'Dongyang Liu', 'Yuewen Cao', 'Xiangchao Yan', 'Xin Li', 'Botian Shi', 'Tao Chen', 'Zhibo Chen', 'Lei Bai', 'Bo Zhang', 'Peng Gao'] | 2,025 | arXiv.org | 2 | 57 | ['Computer Science'] |
2,504.07095 | Neural Motion Simulator: Pushing the Limit of World Models in
Reinforcement Learning | ['Chenjie Hao', 'Weyl Lu', 'Yifan Xu', 'Yubei Chen'] | ['cs.LG', 'cs.RO'] | An embodied system must not only model the patterns of the external world but
also understand its own motion dynamics. A motion dynamic model is essential
for efficient skill acquisition and effective planning. In this work, we
introduce the neural motion simulator (MoSim), a world model that predicts the
future physic... | 2025-04-09T17:59:32Z | 8 pages (main), 2-page appendix, 8 figures, accepted by CVPR 2025 | null | null | null | null | null | null | null | null | null |
2,504.07199 | SemEval-2025 Task 5: LLMs4Subjects -- LLM-based Automated Subject
Tagging for a National Technical Library's Open-Access Catalog | ["Jennifer D'Souza", 'Sameer Sadruddin', 'Holger Israel', 'Mathias Begoin', 'Diana Slawig'] | ['cs.CL', 'cs.AI', 'cs.DL', 'cs.LG'] | We present SemEval-2025 Task 5: LLMs4Subjects, a shared task on automated
subject tagging for scientific and technical records in English and German
using the GND taxonomy. Participants developed LLM-based systems to recommend
top-k subjects, evaluated through quantitative metrics (precision, recall,
F1-score) and qual... | 2025-04-09T18:26:46Z | 10 pages, 4 figures, Accepted as SemEval 2025 Task 5 description
paper | null | null | null | null | null | null | null | null | null |
2,504.0721 | MESA: Text-Driven Terrain Generation Using Latent Diffusion and Global
Copernicus Data | ['Paul Borne--Pons', 'Mikolaj Czerkawski', 'Rosalie Martin', 'Romain Rouffet'] | ['cs.GR', 'cs.CV', 'cs.LG'] | Terrain modeling has traditionally relied on procedural techniques, which
often require extensive domain expertise and handcrafted rules. In this paper,
we present MESA - a novel data-centric alternative by training a diffusion
model on global remote sensing data. This approach leverages large-scale
geospatial informat... | 2025-04-09T18:37:24Z | Accepted at CVPR 2025 Workshop MORSE | null | null | null | null | null | null | null | null | null |
2,504.07448 | LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation | ['Juzheng Zhang', 'Jiacheng You', 'Ashwinee Panda', 'Tom Goldstein'] | ['cs.LG', 'cs.AI', 'cs.CL'] | Low-Rank Adaptation (LoRA) has emerged as a popular parameter-efficient
fine-tuning (PEFT) method for Large Language Models (LLMs), yet it still incurs
notable overhead and suffers from parameter interference in multi-task
scenarios. We propose LoRA with Reduced Interference (LoRI), a simple yet
effective approach that... | 2025-04-10T04:46:04Z | 24 pages, 7 figures, 20 tables | null | null | null | null | null | null | null | null | null |
2,504.07491 | Kimi-VL Technical Report | ['Kimi Team', 'Angang Du', 'Bohong Yin', 'Bowei Xing', 'Bowen Qu', 'Bowen Wang', 'Cheng Chen', 'Chenlin Zhang', 'Chenzhuang Du', 'Chu Wei', 'Congcong Wang', 'Dehao Zhang', 'Dikang Du', 'Dongliang Wang', 'Enming Yuan', 'Enzhe Lu', 'Fang Li', 'Flood Sung', 'Guangda Wei', 'Guokun Lai', 'Han Zhu', 'Hao Ding', 'Hao Hu', 'Ha... | ['cs.CV'] | We present Kimi-VL, an efficient open-source Mixture-of-Experts (MoE)
vision-language model (VLM) that offers advanced multimodal reasoning,
long-context understanding, and strong agent capabilities - all while
activating only 2.8B parameters in its language decoder (Kimi-VL-A3B). Kimi-VL
demonstrates strong performanc... | 2025-04-10T06:48:26Z | Updated Kimi-VL-A3B-Thinking-2506 information | null | null | null | null | null | null | null | null | null |
2,504.07615 | VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model | ['Haozhan Shen', 'Peng Liu', 'Jingcheng Li', 'Chunxin Fang', 'Yibo Ma', 'Jiajia Liao', 'Qiaoli Shen', 'Zilun Zhang', 'Kangjia Zhao', 'Qianqian Zhang', 'Ruochen Xu', 'Tiancheng Zhao'] | ['cs.CV', 'cs.CL'] | Recently DeepSeek R1 has shown that reinforcement learning (RL) can
substantially improve the reasoning capabilities of Large Language Models
(LLMs) through a simple yet effective design. The core of R1 lies in its
rule-based reward formulation, which leverages tasks with deterministic
ground-truth answers to enable pr... | 2025-04-10T10:05:15Z | 11 pages, fix some minor typos in the previous version | null | null | VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model | ['Haozhan Shen', 'Peng Liu', 'Jingcheng Li', 'Chunxin Fang', 'Yibo Ma', 'Jiajia Liao', 'Qiaoli Shen', 'Zilun Zhang', 'Kangjia Zhao', 'Qianqian Zhang', 'Ruochen Xu', 'Tiancheng Zhao'] | 2,025 | arXiv.org | 115 | 58 | ['Computer Science'] |
2,504.07744 | MMLA: Multi-Environment, Multi-Species, Low-Altitude Drone Dataset | ['Jenna Kline', 'Samuel Stevens', 'Guy Maalouf', 'Camille Rondeau Saint-Jean', 'Dat Nguyen Ngoc', 'Majid Mirmehdi', 'David Guerin', 'Tilo Burghardt', 'Elzbieta Pastucha', 'Blair Costelloe', 'Matthew Watson', 'Thomas Richardson', 'Ulrik Pagh Schultz Lundquist'] | ['cs.CV'] | Real-time wildlife detection in drone imagery supports critical ecological
and conservation monitoring. However, standard detection models like YOLO often
fail to generalize across locations and struggle with rare species, limiting
their use in automated drone deployments. We present MMLA, a novel
multi-environment, mu... | 2025-04-10T13:40:27Z | Accepted at CVPR Workshop, CV4Animals 2025 | null | null | null | null | null | null | null | null | null |
2,504.07854 | The KL3M Data Project: Copyright-Clean Training Resources for Large
Language Models | ['Michael J Bommarito II', 'Jillian Bommarito', 'Daniel Martin Katz'] | ['cs.CL', 'cs.AI'] | Practically all large language models have been pre-trained on data that is
subject to global uncertainty related to copyright infringement and breach of
contract. This creates potential risk for users and developers due to this
uncertain legal status. The KL3M Data Project directly confronts this critical
issue by int... | 2025-04-10T15:31:17Z | 27 pages, 7 figures, 9 table | null | null | null | null | null | null | null | null | null |
2,504.07934 | SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual
Reasoning Self-Improvement | ['Xiyao Wang', 'Zhengyuan Yang', 'Chao Feng', 'Hongjin Lu', 'Linjie Li', 'Chung-Ching Lin', 'Kevin Lin', 'Furong Huang', 'Lijuan Wang'] | ['cs.CV'] | We introduce ThinkLite-VL, a family of visual reasoning models that achieve
state-of-the-art (SoTA) performance using an order of magnitude fewer training
samples, relying purely on reinforcement fine-tuning (RFT) self-improvement
without any knowledge distillation. Our central insight is that sample
difficulty critica... | 2025-04-10T17:49:05Z | 27 pages, 5 figures | null | null | SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement | ['Xiyao Wang', 'Zhengyuan Yang', 'Chao Feng', 'Hongjin Lu', 'Linjie Li', 'Chung-Ching Lin', 'K. Lin', 'Furong Huang', 'Lijuan Wang'] | 2,025 | arXiv.org | 19 | 106 | ['Computer Science'] |
2,504.0796 | VisualCloze: A Universal Image Generation Framework via Visual
In-Context Learning | ['Zhong-Yu Li', 'Ruoyi Du', 'Juncheng Yan', 'Le Zhuo', 'Zhen Li', 'Peng Gao', 'Zhanyu Ma', 'Ming-Ming Cheng'] | ['cs.CV'] | Recent progress in diffusion models significantly advances various image
generation tasks. However, the current mainstream approach remains focused on
building task-specific models, which have limited efficiency when supporting a
wide range of different needs. While universal models attempt to address this
limitation, ... | 2025-04-10T17:59:42Z | Project page: https://visualcloze.github.io/ | null | null | null | null | null | null | null | null | null |
2,504.07962 | GLUS: Global-Local Reasoning Unified into A Single Large Language Model
for Video Segmentation | ['Lang Lin', 'Xueyang Yu', 'Ziqi Pang', 'Yu-Xiong Wang'] | ['cs.CV'] | This paper proposes a novel framework utilizing multi-modal large language
models (MLLMs) for referring video object segmentation (RefVOS). Previous
MLLM-based methods commonly struggle with the dilemma between "Ref" and "VOS":
they either specialize in understanding a few key frames (global reasoning) or
tracking obje... | 2025-04-10T17:59:55Z | CVPR 2025 | null | null | null | null | null | null | null | null | null |
2,504.07963 | PixelFlow: Pixel-Space Generative Models with Flow | ['Shoufa Chen', 'Chongjian Ge', 'Shilong Zhang', 'Peize Sun', 'Ping Luo'] | ['cs.CV'] | We present PixelFlow, a family of image generation models that operate
directly in the raw pixel space, in contrast to the predominant latent-space
models. This approach simplifies the image generation process by eliminating
the need for a pre-trained Variational Autoencoder (VAE) and enabling the whole
model end-to-en... | 2025-04-10T17:59:56Z | Technical report. Code: https://github.com/ShoufaChen/PixelFlow | null | null | null | null | null | null | null | null | null |
2,504.07981 | ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer
Use | ['Kaixin Li', 'Ziyang Meng', 'Hongzhan Lin', 'Ziyang Luo', 'Yuchen Tian', 'Jing Ma', 'Zhiyong Huang', 'Tat-Seng Chua'] | ['cs.CV', 'cs.HC', 'cs.MM', '68-11 68-04', 'I.2.7; I.2.10'] | Recent advancements in Multi-modal Large Language Models (MLLMs) have led to
significant progress in developing GUI agents for general tasks such as web
browsing and mobile phone use. However, their application in professional
domains remains under-explored. These specialized workflows introduce unique
challenges for G... | 2025-04-04T14:25:17Z | 13pages | null | null | null | null | null | null | null | null | null |
2,504.08016 | Emergence of psychopathological computations in large language models | ['Soo Yong Lee', 'Hyunjin Hwang', 'Taekwan Kim', 'Yuyeong Kim', 'Kyuri Park', 'Jaemin Yoo', 'Denny Borsboom', 'Kijung Shin'] | ['q-bio.NC', 'cs.AI', 'cs.CL'] | Can large language models (LLMs) implement computations of psychopathology?
An effective approach to the question hinges on addressing two factors. First,
for conceptual validity, we require a general and computational account of
psychopathology that is applicable to computational entities without biological
embodiment... | 2025-04-10T15:36:30Z | pre-print | null | null | null | null | null | null | null | null | null |
2,504.08548 | COP-GEN-Beta: Unified Generative Modelling of COPernicus Imagery
Thumbnails | ['Miguel Espinosa', 'Valerio Marsocci', 'Yuru Jia', 'Elliot J. Crowley', 'Mikolaj Czerkawski'] | ['cs.GR', 'cs.CV'] | In remote sensing, multi-modal data from various sensors capturing the same
scene offers rich opportunities, but learning a unified representation across
these modalities remains a significant challenge. Traditional methods have
often been limited to single or dual-modality approaches. In this paper, we
introduce COP-G... | 2025-04-11T14:00:46Z | Accepted at CVPR 2025 Workshop MORSE | null | null | null | null | null | null | null | null | null |
2,504.08596 | MedHal: An Evaluation Dataset for Medical Hallucination Detection | ['Gaya Mehenni', 'Amal Zouaq'] | ['cs.CL', 'cs.AI', 'I.2.7'] | We present MedHal, a novel large-scale dataset specifically designed to
evaluate if models can detect hallucinations in medical texts. Current
hallucination detection methods face significant limitations when applied to
specialized domains like medicine, where they can have disastrous consequences.
Existing medical dat... | 2025-04-11T14:55:15Z | null | null | null | MedHal: An Evaluation Dataset for Medical Hallucination Detection | ['Gaya Mehenni', 'Amal Zouaq'] | 2,025 | arXiv.org | 0 | 21 | ['Computer Science'] |
2,504.086 | SQL-R1: Training Natural Language to SQL Reasoning Model By
Reinforcement Learning | ['Peixian Ma', 'Xialie Zhuang', 'Chengjin Xu', 'Xuhui Jiang', 'Ran Chen', 'Jian Guo'] | ['cs.DB'] | Natural Language to SQL (NL2SQL) enables intuitive interactions with
databases by transforming natural language queries into structured SQL
statements. Despite recent advancements in enhancing human-computer interaction
within database applications, significant challenges persist, particularly
regarding the inference p... | 2025-04-11T15:01:30Z | null | null | null | null | null | null | null | null | null | null |
2,504.08672 | Genius: A Generalizable and Purely Unsupervised Self-Training Framework
For Advanced Reasoning | ['Fangzhi Xu', 'Hang Yan', 'Chang Ma', 'Haiteng Zhao', 'Qiushi Sun', 'Kanzhi Cheng', 'Junxian He', 'Jun Liu', 'Zhiyong Wu'] | ['cs.CL', 'cs.AI', 'cs.LG'] | Advancing LLM reasoning skills has captivated wide interest. However, current
post-training techniques rely heavily on supervisory signals, such as outcome
supervision or auxiliary reward models, which face the problem of scalability
and high annotation costs. This motivates us to enhance LLM reasoning without
the need... | 2025-04-11T16:26:23Z | 14 pages, 7 figures | null | null | Genius: A Generalizable and Purely Unsupervised Self-Training Framework For Advanced Reasoning | ['Fangzhi Xu', 'Hang Yan', 'Chang Ma', 'Haiteng Zhao', 'Qiushi Sun', 'Kanzhi Cheng', 'Junxian He', 'Jun Liu', 'Zhiyong Wu'] | 2,025 | arXiv.org | 5 | 59 | ['Computer Science'] |
2,504.08716 | ModernBERT or DeBERTaV3? Examining Architecture and Data Influence on
Transformer Encoder Models Performance | ['Wissam Antoun', 'Benoît Sagot', 'Djamé Seddah'] | ['cs.CL'] | Pretrained transformer-encoder models like DeBERTaV3 and ModernBERT introduce
architectural advancements aimed at improving efficiency and performance.
Although the authors of ModernBERT report improved performance over DeBERTaV3
on several benchmarks, the lack of disclosed training data and the absence of
comparisons ... | 2025-04-11T17:29:35Z | Preprint. Under review | null | null | null | null | null | null | null | null | null |
2,504.08837 | VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models
with Reinforcement Learning | ['Haozhe Wang', 'Chao Qu', 'Zuming Huang', 'Wei Chu', 'Fangzhen Lin', 'Wenhu Chen'] | ['cs.LG', 'cs.AI'] | Recently, slow-thinking systems like GPT-o1 and DeepSeek-R1 have demonstrated
great potential in solving challenging problems through explicit reflection.
They significantly outperform the best fast-thinking models, such as GPT-4o, on
various math and science benchmarks. However, their multimodal reasoning
capabilities... | 2025-04-10T17:41:56Z | Preprint | null | null | null | null | null | null | null | null | null |
2,504.08846 | AI-University: An LLM-based platform for instructional alignment to
scientific classrooms | ['Mostafa Faghih Shojaei', 'Rahul Gulati', 'Benjamin A. Jasperson', 'Shangshang Wang', 'Simone Cimolato', 'Dangli Cao', 'Willie Neiswanger', 'Krishna Garikipati'] | ['cs.CY', 'cs.AI', 'cs.CL', 'cs.LG'] | We introduce AI University (AI-U), a flexible framework for AI-driven course
content delivery that adapts to instructors' teaching styles. At its core, AI-U
fine-tunes a large language model (LLM) with retrieval-augmented generation
(RAG) to generate instructor-aligned responses from lecture videos, notes, and
textbook... | 2025-04-11T01:26:34Z | 10 pages, 3 figures | null | null | null | null | null | null | null | null | null |
2,504.09184 | Parameterized Synthetic Text Generation with SimpleStories | ['Lennart Finke', 'Chandan Sreedhara', 'Thomas Dooms', 'Mat Allen', 'Emerald Zhang', 'Juan Diego Rodriguez', 'Noa Nabeshima', 'Thomas Marshall', 'Dan Braun'] | ['cs.CL', 'cs.AI'] | We present SimpleStories, a large synthetic story dataset in simple language,
consisting of 2 million samples each in English and Japanese. Through
parameterizing prompts at multiple levels of abstraction, we achieve control
over story characteristics at scale, inducing syntactic and semantic diversity.
Ablations on a ... | 2025-04-12T11:44:47Z | null | null | null | null | null | null | null | null | null | null |
2,504.09228 | Learning Occlusion-Robust Vision Transformers for Real-Time UAV Tracking | ['You Wu', 'Xucheng Wang', 'Xiangyang Yang', 'Mengyuan Liu', 'Dan Zeng', 'Hengzhou Ye', 'Shuiwang Li'] | ['cs.CV'] | Single-stream architectures using Vision Transformer (ViT) backbones show
great potential for real-time UAV tracking recently. However, frequent
occlusions from obstacles like buildings and trees expose a major drawback:
these models often lack strategies to handle occlusions effectively. New
methods are needed to enha... | 2025-04-12T14:06:50Z | null | null | null | null | null | null | null | null | null | null |
2,504.09421 | ClinicalGPT-R1: Pushing reasoning capability of generalist disease
diagnosis with large language model | ['Wuyang Lan', 'Wenzheng Wang', 'Changwei Ji', 'Guoxing Yang', 'Yongbo Zhang', 'Xiaohong Liu', 'Song Wu', 'Guangyu Wang'] | ['cs.CL', 'cs.AI'] | Recent advances in reasoning with large language models (LLMs)has shown
remarkable reasoning capabilities in domains such as mathematics and coding,
yet their application to clinical diagnosis remains underexplored. Here, we
introduce ClinicalGPT-R1, a reasoning enhanced generalist large language model
for disease diag... | 2025-04-13T04:00:40Z | 8 pages, 6 figures | null | null | ClinicalGPT-R1: Pushing reasoning capability of generalist disease diagnosis with large language model | ['Wuyang Lan', 'Wenzheng Wang', 'Changwei Ji', 'Guoxing Yang', 'Yongbo Zhang', 'Xiaohong Liu', 'Song Wu', 'Guangyu Wang'] | 2,025 | arXiv.org | 3 | 23 | ['Computer Science'] |
2,504.0957 | LLMs Can Achieve High-quality Simultaneous Machine Translation as
Efficiently as Offline | ['Biao Fu', 'Minpeng Liao', 'Kai Fan', 'Chengxi Li', 'Liang Zhang', 'Yidong Chen', 'Xiaodong Shi'] | ['cs.CL'] | When the complete source sentence is provided, Large Language Models (LLMs)
perform excellently in offline machine translation even with a simple prompt
"Translate the following sentence from [src lang] into [tgt lang]:". However,
in many real scenarios, the source tokens arrive in a streaming manner and
simultaneous m... | 2025-04-13T13:45:53Z | Camera ready version for ACL 2025 Findings | null | null | LLMs Can Achieve High-quality Simultaneous Machine Translation as Efficiently as Offline | ['Biao Fu', 'Minpeng Liao', 'Kai Fan', 'Chengxi Li', 'Liang Zhang', 'Yidong Chen', 'Xiaodong Shi'] | 2,025 | arXiv.org | 1 | 40 | ['Computer Science'] |
2,504.09641 | TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning | ['Xingjian Zhang', 'Siwei Wen', 'Wenjun Wu', 'Lei Huang'] | ['cs.CV'] | Recently, improving the reasoning ability of large multimodal models (LMMs)
through reinforcement learning has made great progress. However, most existing
works are based on highly reasoning-intensive datasets such as mathematics and
code, and researchers generally choose large-scale models as the foundation. We
argue ... | 2025-04-13T16:32:49Z | null | null | null | TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning | ['Xingjian Zhang', 'Siwei Wen', 'Wenjun Wu', 'Lei Huang'] | 2,025 | arXiv.org | 16 | 29 | ['Computer Science'] |
2,504.09645 | Myanmar XNLI: Building a Dataset and Exploring Low-resource Approaches
to Natural Language Inference with Myanmar | ['Aung Kyaw Htet', 'Mark Dras'] | ['cs.CL', 'cs.AI'] | Despite dramatic recent progress in NLP, it is still a major challenge to
apply Large Language Models (LLM) to low-resource languages. This is made
visible in benchmarks such as Cross-Lingual Natural Language Inference (XNLI),
a key task that demonstrates cross-lingual capabilities of NLP systems across a
set of 15 lan... | 2025-04-13T16:36:59Z | null | null | null | Myanmar XNLI: Building a Dataset and Exploring Low-resource Approaches to Natural Language Inference with Myanmar | ['Kyaw Htet Aung', 'M. Dras'] | 2,025 | Language Resources and Evaluation | 1 | 31 | ['Computer Science'] |
2,504.09696 | GRPO-LEAD: A Difficulty-Aware Reinforcement Learning Approach for
Concise Mathematical Reasoning in Language Models | ['Jixiao Zhang', 'Chunsheng Zuo'] | ['cs.CL'] | Recent advances in R1-like reasoning models leveraging Group Relative Policy
Optimization (GRPO) have significantly improved the performance of language
models on mathematical reasoning tasks. However, current GRPO implementations
encounter critical challenges, including reward sparsity due to binary accuracy
metrics, ... | 2025-04-13T19:07:45Z | null | null | null | null | null | null | null | null | null | null |
2,504.09697 | SPICE: A Synergistic, Precise, Iterative, and Customizable Image Editing
Workflow | ['Kenan Tang', 'Yanhong Li', 'Yao Qin'] | ['cs.GR', 'cs.CV', 'cs.LG'] | Recent prompt-based image editing models have demonstrated impressive
prompt-following capability at structural editing tasks. However, existing
models still fail to perform local edits, follow detailed editing prompts, or
maintain global image quality beyond a single editing step. To address these
challenges, we intro... | 2025-04-13T19:13:04Z | 24 pages, 21 figures. Figure 9(b) has been accepted by CVPR AI Art
Gallery 2025 | null | null | SPICE: A Synergistic, Precise, Iterative, and Customizable Image Editing Workflow | ['Kenan Tang', 'Yanhong Li', 'Yao Qin'] | 2,025 | arXiv.org | 0 | 30 | ['Computer Science'] |
2,504.09763 | Executable Functional Abstractions: Inferring Generative Programs for
Advanced Math Problems | ['Zaid Khan', 'Elias Stengel-Eskin', 'Archiki Prasad', 'Jaemin Cho', 'Mohit Bansal'] | ['cs.CL', 'cs.AI', 'cs.LG'] | Scientists often infer abstract procedures from specific instances of
problems and use the abstractions to generate new, related instances. For
example, programs encoding the formal rules and properties of a system have
been useful in fields ranging from RL (procedural environments) to physics
(simulation engines). The... | 2025-04-14T00:06:48Z | Project Page: https://zaidkhan.me/EFAGen/ | null | null | null | null | null | null | null | null | null |
2,504.09772 | Two Heads are Better Than One: Test-time Scaling of Multi-agent
Collaborative Reasoning | ['Can Jin', 'Hongwu Peng', 'Qixin Zhang', 'Yujin Tang', 'Dimitris N. Metaxas', 'Tong Che'] | ['cs.AI'] | Multi-agent systems (MAS) built on large language models (LLMs) offer a
promising path toward solving complex, real-world tasks that single-agent
systems often struggle to manage. While recent advancements in test-time
scaling (TTS) have significantly improved single-agent performance on
challenging reasoning tasks, ho... | 2025-04-14T00:27:45Z | null | null | null | null | null | null | null | null | null | null |
2,504.09795 | VDocRAG: Retrieval-Augmented Generation over Visually-Rich Documents | ['Ryota Tanaka', 'Taichi Iki', 'Taku Hasegawa', 'Kyosuke Nishida', 'Kuniko Saito', 'Jun Suzuki'] | ['cs.CL', 'cs.AI', 'cs.CV', 'cs.IR'] | We aim to develop a retrieval-augmented generation (RAG) framework that
answers questions over a corpus of visually-rich documents presented in mixed
modalities (e.g., charts, tables) and diverse formats (e.g., PDF, PPTX). In
this paper, we introduce a new RAG framework, VDocRAG, which can directly
understand varied do... | 2025-04-14T01:50:33Z | Accepted by CVPR 2025; project page: https://vdocrag.github.io | null | null | null | null | null | null | null | null | null |
2,504.09881 | Focus on Local: Finding Reliable Discriminative Regions for Visual Place
Recognition | ['Changwei Wang', 'Shunpeng Chen', 'Yukun Song', 'Rongtao Xu', 'Zherui Zhang', 'Jiguang Zhang', 'Haoran Yang', 'Yu Zhang', 'Kexue Fu', 'Shide Du', 'Zhiwei Xu', 'Longxiang Gao', 'Li Guo', 'Shibiao Xu'] | ['cs.CV'] | Visual Place Recognition (VPR) is aimed at predicting the location of a query
image by referencing a database of geotagged images. For VPR task, often fewer
discriminative local regions in an image produce important effects while
mundane background regions do not contribute or even cause perceptual aliasing
because of ... | 2025-04-14T05:04:51Z | Accepted by AAAI 2025 | null | null | null | null | null | null | null | null | null |
2,504.09887 | Enhanced Semantic Extraction and Guidance for UGC Image Super Resolution | ['Yiwen Wang', 'Ying Liang', 'Yuxuan Zhang', 'Xinning Chai', 'Zhengxue Cheng', 'Yingsheng Qin', 'Yucai Yang', 'Rong Xie', 'Li Song'] | ['cs.CV'] | Due to the disparity between real-world degradations in user-generated
content(UGC) images and synthetic degradations, traditional super-resolution
methods struggle to generalize effectively, necessitating a more robust
approach to model real-world distortions. In this paper, we propose a novel
approach to UGC image su... | 2025-04-14T05:26:24Z | null | null | null | null | null | null | null | null | null | null |
2,504.09925 | FUSION: Fully Integration of Vision-Language Representations for Deep
Cross-Modal Understanding | ['Zheng Liu', 'Mengjie Liu', 'Jingzhou Chen', 'Jingwei Xu', 'Bin Cui', 'Conghui He', 'Wentao Zhang'] | ['cs.CV'] | We introduce FUSION, a family of multimodal large language models (MLLMs)
with a fully vision-language alignment and integration paradigm. Unlike
existing methods that primarily rely on late-stage modality interaction during
LLM decoding, our approach achieves deep, dynamic integration throughout the
entire processing ... | 2025-04-14T06:33:29Z | null | null | null | null | null | null | null | null | null | null |
2,504.09975 | OctGPT: Octree-based Multiscale Autoregressive Models for 3D Shape
Generation | ['Si-Tong Wei', 'Rui-Huan Wang', 'Chuan-Zhi Zhou', 'Baoquan Chen', 'Peng-Shuai Wang'] | ['cs.GR', 'cs.CV'] | Autoregressive models have achieved remarkable success across various
domains, yet their performance in 3D shape generation lags significantly behind
that of diffusion models. In this paper, we introduce OctGPT, a novel
multiscale autoregressive model for 3D shape generation that dramatically
improves the efficiency an... | 2025-04-14T08:31:26Z | SIGGRAPH 2025 | null | null | null | null | null | null | null | null | null |
2,504.10044 | Aligning Anime Video Generation with Human Feedback | ['Bingwen Zhu', 'Yudong Jiang', 'Baohan Xu', 'Siqian Yang', 'Mingyu Yin', 'Yidi Wu', 'Huyang Sun', 'Zuxuan Wu'] | ['cs.CV'] | Anime video generation faces significant challenges due to the scarcity of
anime data and unusual motion patterns, leading to issues such as motion
distortion and flickering artifacts, which result in misalignment with human
preferences. Existing reward models, designed primarily for real-world videos,
fail to capture ... | 2025-04-14T09:49:34Z | 10 pages, 7 figures, 7 tables | null | null | Aligning Anime Video Generation with Human Feedback | ['Bingwen Zhu', 'Yudong Jiang', 'Baohan Xu', 'Siqian Yang', 'Mingyu Yin', 'Yidi Wu', 'Huyang Sun', 'Zuxuan Wu'] | 2,025 | arXiv.org | 0 | 44 | ['Computer Science'] |
2,504.10449 | M1: Towards Scalable Test-Time Compute with Mamba Reasoning Models | ['Junxiong Wang', 'Wen-Ding Li', 'Daniele Paliotta', 'Daniel Ritter', 'Alexander M. Rush', 'Tri Dao'] | ['cs.LG'] | Effective reasoning is crucial to solving complex mathematical problems.
Recent large language models (LLMs) have boosted performance by scaling
test-time computation through long chain-of-thought reasoning. However,
transformer-based models are inherently limited in extending context length due
to their quadratic comp... | 2025-04-14T17:38:25Z | Code is available https://github.com/jxiw/M1 | null | null | null | null | null | null | null | null | null |
2,504.10458 | GUI-R1 : A Generalist R1-Style Vision-Language Action Model For GUI
Agents | ['Run Luo', 'Lu Wang', 'Wanwei He', 'Xiaobo Xia'] | ['cs.CV', 'cs.CL', 'cs.HC'] | Existing efforts in building Graphical User Interface (GUI) agents largely
rely on the training paradigm of supervised fine-tuning on Large
Vision-Language Models (LVLMs). However, this approach not only demands
extensive amounts of training data but also struggles to effectively understand
GUI screenshots and generali... | 2025-04-14T17:45:54Z | null | null | null | null | null | null | null | null | null | null |
2,504.10462 | The Scalability of Simplicity: Empirical Analysis of Vision-Language
Learning with a Single Transformer | ['Weixian Lei', 'Jiacong Wang', 'Haochen Wang', 'Xiangtai Li', 'Jun Hao Liew', 'Jiashi Feng', 'Zilong Huang'] | ['cs.CV'] | This paper introduces SAIL, a single transformer unified multimodal large
language model (MLLM) that integrates raw pixel encoding and language decoding
within a singular architecture. Unlike existing modular MLLMs, which rely on a
pre-trained vision transformer (ViT), SAIL eliminates the need for a separate
vision enc... | 2025-04-14T17:50:20Z | null | null | null | null | null | null | null | null | null | null |
2,504.10479 | InternVL3: Exploring Advanced Training and Test-Time Recipes for
Open-Source Multimodal Models | ['Jinguo Zhu', 'Weiyun Wang', 'Zhe Chen', 'Zhaoyang Liu', 'Shenglong Ye', 'Lixin Gu', 'Hao Tian', 'Yuchen Duan', 'Weijie Su', 'Jie Shao', 'Zhangwei Gao', 'Erfei Cui', 'Xuehui Wang', 'Yue Cao', 'Yangzhou Liu', 'Xingguang Wei', 'Hongjie Zhang', 'Haomin Wang', 'Weiye Xu', 'Hao Li', 'Jiahao Wang', 'Nianchen Deng', 'Songze ... | ['cs.CV'] | We introduce InternVL3, a significant advancement in the InternVL series
featuring a native multimodal pre-training paradigm. Rather than adapting a
text-only large language model (LLM) into a multimodal large language model
(MLLM) that supports visual inputs, InternVL3 jointly acquires multimodal and
linguistic capabi... | 2025-04-14T17:59:25Z | Technical Report | null | null | InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models | ['Jinguo Zhu', 'Weiyun Wang', 'Zhe Chen', 'Zhaoyang Liu', 'Shenglong Ye', 'Lixin Gu', 'Yuchen Duan', 'Hao Tian', 'Weijie Su', 'Jie Shao', 'Zhangwei Gao', 'Erfei Cui', 'Yue Cao', 'Yangzhou Liu', 'Haomin Wang', 'Weiye Xu', 'Hao Li', 'Jiahao Wang', 'Han Lv', 'Dengnian Chen', 'Songze Li', 'Yinan He', 'Tan Jiang', 'Jiapeng ... | 2,025 | arXiv.org | 132 | 150 | ['Computer Science'] |
2,504.10481 | xVerify: Efficient Answer Verifier for Reasoning Model Evaluations | ['Ding Chen', 'Qingchen Yu', 'Pengyuan Wang', 'Wentao Zhang', 'Bo Tang', 'Feiyu Xiong', 'Xinchi Li', 'Minchuan Yang', 'Zhiyu Li'] | ['cs.CL'] | With the release of the o1 model by OpenAI, reasoning models adopting slow
thinking strategies have gradually emerged. As the responses generated by such
models often include complex reasoning, intermediate steps, and
self-reflection, existing evaluation methods are often inadequate. They
struggle to determine whether ... | 2025-04-14T17:59:36Z | 32 pages | null | null | null | null | null | null | null | null | null |
2,504.10483 | REPA-E: Unlocking VAE for End-to-End Tuning with Latent Diffusion
Transformers | ['Xingjian Leng', 'Jaskirat Singh', 'Yunzhong Hou', 'Zhenchang Xing', 'Saining Xie', 'Liang Zheng'] | ['cs.CV', 'cs.LG'] | In this paper we tackle a fundamental question: "Can we train latent
diffusion models together with the variational auto-encoder (VAE) tokenizer in
an end-to-end manner?" Traditional deep-learning wisdom dictates that
end-to-end training is often preferable when possible. However, for latent
diffusion transformers, it ... | 2025-04-14T17:59:53Z | null | null | null | REPA-E: Unlocking VAE for End-to-End Tuning with Latent Diffusion Transformers | ['Xingjian Leng', 'Jaskirat Singh', 'Yunzhong Hou', 'Zhenchang Xing', 'Saining Xie', 'Liang Zheng'] | 2,025 | arXiv.org | 6 | 55 | ['Computer Science'] |
2,504.10647 | Improving In-Context Learning with Reasoning Distillation | ['Nafis Sadeq', 'Xin Xu', 'Zhouhang Xie', 'Julian McAuley', 'Byungkyu Kang', 'Prarit Lamba', 'Xiang Gao'] | ['cs.CL'] | Language models rely on semantic priors to perform in-context learning, which
leads to poor performance on tasks involving inductive reasoning.
Instruction-tuning methods based on imitation learning can superficially
enhance the in-context learning performance of language models, but they often
fail to improve the mode... | 2025-04-14T18:59:10Z | null | null | null | null | null | null | null | null | null | null |
2,504.10659 | Relation-Rich Visual Document Generator for Visual Information
Extraction | ['Zi-Han Jiang', 'Chien-Wei Lin', 'Wei-Hua Li', 'Hsuan-Tung Liu', 'Yi-Ren Yeh', 'Chu-Song Chen'] | ['cs.CV'] | Despite advances in Large Language Models (LLMs) and Multimodal LLMs (MLLMs)
for visual document understanding (VDU), visual information extraction (VIE)
from relation-rich documents remains challenging due to the layout diversity
and limited training data. While existing synthetic document generators attempt
to addres... | 2025-04-14T19:19:26Z | CVPR 2025 | null | null | null | null | null | null | null | null | null |
2,504.10686 | The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report | ['Bin Ren', 'Hang Guo', 'Lei Sun', 'Zongwei Wu', 'Radu Timofte', 'Yawei Li', 'Yao Zhang', 'Xinning Chai', 'Zhengxue Cheng', 'Yingsheng Qin', 'Yucai Yang', 'Li Song', 'Hongyuan Yu', 'Pufan Xu', 'Cheng Wan', 'Zhijuan Huang', 'Peng Guo', 'Shuyuan Cui', 'Chenjun Li', 'Xuehai Hu', 'Pan Pan', 'Xin Zhang', 'Heng Zhang', 'Qing... | ['cs.CV', 'eess.IV'] | This paper presents a comprehensive review of the NTIRE 2025 Challenge on
Single-Image Efficient Super-Resolution (ESR). The challenge aimed to advance
the development of deep models that optimize key computational metrics, i.e.,
runtime, parameters, and FLOPs, while achieving a PSNR of at least 26.90 dB on
the $\opera... | 2025-04-14T20:18:21Z | Accepted by CVPR2025 NTIRE Workshop, Efficient Super-Resolution
Challenge Report. 50 pages | null | null | null | null | null | null | null | null | null |
2,504.10694 | The Jailbreak Tax: How Useful are Your Jailbreak Outputs? | ['Kristina Nikolić', 'Luze Sun', 'Jie Zhang', 'Florian Tramèr'] | ['cs.LG', 'cs.AI', 'cs.CR'] | Jailbreak attacks bypass the guardrails of large language models to produce
harmful outputs. In this paper, we ask whether the model outputs produced by
existing jailbreaks are actually useful. For example, when jailbreaking a model
to give instructions for building a bomb, does the jailbreak yield good
instructions? S... | 2025-04-14T20:30:41Z | null | null | null | null | null | null | null | null | null | null |
2,504.11001 | ReZero: Enhancing LLM search ability by trying one-more-time | ['Alan Dao', 'Thinh Le'] | ['cs.CL'] | Retrieval-Augmented Generation (RAG) improves Large Language Model (LLM)
performance on knowledge-intensive tasks but depends heavily on initial search
query quality. Current methods, often using Reinforcement Learning (RL),
typically focus on query formulation or reasoning over results, without
explicitly encouraging ... | 2025-04-15T09:18:21Z | null | null | null | null | null | null | null | null | null | null |
2,504.11109 | Fine-Tuning Large Language Models on Quantum Optimization Problems for
Circuit Generation | ['Linus Jern', 'Valter Uotila', 'Cong Yu', 'Bo Zhao'] | ['quant-ph', 'cs.AI'] | Large language models (LLM) have achieved remarkable outcomes in addressing
complex problems, including math, coding, and analyzing large amounts of
scientific reports. Yet few works have explored the potential of LLM in quantum
computing. The most challenging problem is how to leverage LLMs to
automatically generate q... | 2025-04-15T11:56:54Z | 12 pages, 8 figures, 3 tables | null | null | null | null | null | null | null | null | null |
2,504.11171 | TerraMind: Large-Scale Generative Multimodality for Earth Observation | ['Johannes Jakubik', 'Felix Yang', 'Benedikt Blumenstiel', 'Erik Scheurer', 'Rocco Sedona', 'Stefano Maurogiovanni', 'Jente Bosmans', 'Nikolaos Dionelis', 'Valerio Marsocci', 'Niklas Kopp', 'Rahul Ramachandran', 'Paolo Fraccaro', 'Thomas Brunschwiler', 'Gabriele Cavallaro', 'Juan Bernabe-Moreno', 'Nicolas Longépé'] | ['cs.CV', 'cs.AI'] | We present TerraMind, the first any-to-any generative, multimodal foundation
model for Earth observation (EO). Unlike other multimodal models, TerraMind is
pretrained on dual-scale representations combining both token-level and
pixel-level data across modalities. On a token level, TerraMind encodes
high-level contextua... | 2025-04-15T13:17:39Z | Accepted at ICCV'25 | null | null | null | null | null | null | null | null | null |
2,504.11271 | Distillation-Supervised Convolutional Low-Rank Adaptation for Efficient
Image Super-Resolution | ['Xinning Chai', 'Yao Zhang', 'Yuxuan Zhang', 'Zhengxue Cheng', 'Yingsheng Qin', 'Yucai Yang', 'Li Song'] | ['cs.CV'] | Convolutional neural networks (CNNs) have been widely used in efficient image
super-resolution. However, for CNN-based methods, performance gains often
require deeper networks and larger feature maps, which increase complexity and
inference costs. Inspired by LoRA's success in fine-tuning large language
models, we expl... | 2025-04-15T15:12:57Z | null | null | null | Distillation-Supervised Convolutional Low-Rank Adaptation for Efficient Image Super-Resolution | ['Xinning Chai', 'Yao Zhang', 'Yuxuan Zhang', 'Zhengxue Cheng', 'Yingsheng Qin', 'Yucai Yang', 'Li Song'] | 2,025 | arXiv.org | 0 | 0 | ['Computer Science'] |
2,504.11289 | UniAnimate-DiT: Human Image Animation with Large-Scale Video Diffusion
Transformer | ['Xiang Wang', 'Shiwei Zhang', 'Longxiang Tang', 'Yingya Zhang', 'Changxin Gao', 'Yuehuan Wang', 'Nong Sang'] | ['cs.CV'] | This report presents UniAnimate-DiT, an advanced project that leverages the
cutting-edge and powerful capabilities of the open-source Wan2.1 model for
consistent human image animation. Specifically, to preserve the robust
generative capabilities of the original Wan2.1 model, we implement Low-Rank
Adaptation (LoRA) tech... | 2025-04-15T15:29:11Z | The training and inference code (based on Wan2.1) is available at
https://github.com/ali-vilab/UniAnimate-DiT | null | null | null | null | null | null | null | null | null |
2,504.11343 | A Minimalist Approach to LLM Reasoning: from Rejection Sampling to
Reinforce | ['Wei Xiong', 'Jiarui Yao', 'Yuhui Xu', 'Bo Pang', 'Lei Wang', 'Doyen Sahoo', 'Junnan Li', 'Nan Jiang', 'Tong Zhang', 'Caiming Xiong', 'Hanze Dong'] | ['cs.LG', 'cs.AI', 'cs.CL', 'stat.ML'] | Reinforcement learning (RL) has become a prevailing approach for fine-tuning
large language models (LLMs) on complex reasoning tasks. Among recent methods,
GRPO stands out for its empirical success in training models such as
DeepSeek-R1, yet the sources of its effectiveness remain poorly understood. In
this work, we re... | 2025-04-15T16:15:02Z | null | null | null | null | null | null | null | null | null | null |
2,504.11354 | Kimina-Prover Preview: Towards Large Formal Reasoning Models with
Reinforcement Learning | ['Haiming Wang', 'Mert Unsal', 'Xiaohan Lin', 'Mantas Baksys', 'Junqi Liu', 'Marco Dos Santos', 'Flood Sung', 'Marina Vinyes', 'Zhenzhe Ying', 'Zekai Zhu', 'Jianqiao Lu', 'Hugues de Saxcé', 'Bolton Bailey', 'Chendong Song', 'Chenjun Xiao', 'Dehao Zhang', 'Ebony Zhang', 'Frederick Pu', 'Han Zhu', 'Jiawei Liu', 'Jonas Ba... | ['cs.AI'] | We introduce Kimina-Prover Preview, a large language model that pioneers a
novel reasoning-driven exploration paradigm for formal theorem proving, as
showcased in this preview release. Trained with a large-scale reinforcement
learning pipeline from Qwen2.5-72B, Kimina-Prover demonstrates strong
performance in Lean 4 pr... | 2025-04-15T16:23:44Z | 22 pages | null | null | null | null | null | null | null | null | null |
2,504.11456 | DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and
Verifiable Mathematical Dataset for Advancing Reasoning | ['Zhiwei He', 'Tian Liang', 'Jiahao Xu', 'Qiuzhi Liu', 'Xingyu Chen', 'Yue Wang', 'Linfeng Song', 'Dian Yu', 'Zhenwen Liang', 'Wenxuan Wang', 'Zhuosheng Zhang', 'Rui Wang', 'Zhaopeng Tu', 'Haitao Mi', 'Dong Yu'] | ['cs.CL', 'cs.AI'] | Reinforcement learning (RL) with large language models shows promise in
complex reasoning. However, its progress is hindered by the lack of large-scale
training data that is sufficiently challenging, contamination-free and
verifiable. To this end, we introduce DeepMath-103K, a large-scale mathematical
dataset designed ... | 2025-04-15T17:59:51Z | WIP | null | null | null | null | null | null | null | null | null |
2,504.11514 | Enhancing Autonomous Driving Systems with On-Board Deployed Large
Language Models | ['Nicolas Baumann', 'Cheng Hu', 'Paviththiren Sivasothilingam', 'Haotong Qin', 'Lei Xie', 'Michele Magno', 'Luca Benini'] | ['cs.AI', 'cs.RO'] | Neural Networks (NNs) trained through supervised learning struggle with
managing edge-case scenarios common in real-world driving due to the
intractability of exhaustive datasets covering all edge-cases, making
knowledge-driven approaches, akin to how humans intuitively detect unexpected
driving behavior, a suitable co... | 2025-04-15T13:49:17Z | null | null | null | Enhancing Autonomous Driving Systems with On-Board Deployed Large Language Models | ['Nicolas Baumann', 'Cheng Hu', 'Paviththiren Sivasothilingam', 'Haotong Qin', 'Lei Xie', 'Michele Magno', 'Luca Benini'] | 2,025 | arXiv.org | 2 | 33 | ['Computer Science'] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.