Create testreadme.md

testreadme.md

@@ -0,0 +1,86 @@
+**模型相关论文可查阅:**
+
+**7B**
+
+[Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration](https://arxiv.org/abs/2508.13755) <br>
+[TiFRe: Text-guided Video Frame Reduction for Efficient Video Multi-modal Large Language Models](https://arxiv.org/abs/2602.08861) <br>
+[Searching Meta Reasoning Skeleton to Guide LLM Reasoning](https://arxiv.org/abs/2510.04116) <br>
+[Post-Training Quantization of OpenPangu Models for Efficient Deployment on Atlas A2](https://arxiv.org/abs/2512.23367) <br>
+[OSUM-Pangu: An Open-Source Multidimension Speech Understanding Foundation Model Built upon OpenPangu on Ascend NPUs](https://arxiv.org/abs/2603.10862) <br>
+[Structured Episodic Event Memory](https://arxiv.org/abs/2601.06411) <br>
+[Quasar: Quantized Self-Speculative Acceleration for Rapid Inference via Memory-Efficient Verification](https://arxiv.org/abs/2603.01399v1) <br>
+[CSSBench: Evaluating the Safety of Lightweight LLMs against Chinese-Specific Adversarial Patterns](https://arxiv.org/abs/2601.00588) <br>
+[Beyond Static Question Banks: Dynamic Knowledge Expansion via LLM-Automated Graph Construction and Adaptive Generation](https://arxiv.org/abs/2602.00020) <br>
+[Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking](https://arxiv.org/abs/2509.23392v3) <br>
+[EvoOpt-LLM: Evolving industrial optimization models with large language models](https://arxiv.org/abs/2602.01082) <br>
+[Accelerating OpenPangu Inference on NPU via Speculative Decoding](https://arxiv.org/abs/2603.03383) <br>
+[V-CAGE: Context-Aware Generation and Verification for Scalable Long-Horizon Embodied Tasks](https://arxiv.org/abs/2601.15164) <br>
+[Scheduling LLM Inference with Uncertainty-Aware Output Length Predictions](https://arxiv.org/abs/2604.00499) <br>
+[Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Prompt](https://arxiv.org/abs/2602.11513) <br>
+[A-IO: Adaptive Inference Orchestration for Memory-Bound NPUs](https://arxiv.org/abs/2604.09752) <br>
+[Multi-source Heterogeneous Public Opinion Analysis via Collaborative Reasoning and Adaptive Fusion: A Systematically Integrated Approach](https://arxiv.org/abs/2602.15857) <br>
+[Characterize LSM-tree Compaction Performance via On-Device LLM Inference](https://arxiv.org/abs/2602.12669v1) <br>
+[Traceable Cross-Source RAG for Chinese Tibetan Medicine Question Answering](https://arxiv.org/abs/2602.05195) <br>
+[A Comprehensive Evaluation of LLM Reasoning: From Single-Model to Multi-Agent Paradigms](https://arxiv.org/abs/2601.13243) <br>
+[Structured Self-Consistency: A Multi-Task Evaluation of LLMs on VirtualHome](https://arxiv.org/abs/2602.00611) <br>
+[ArkEval: Benchmarking and Evaluating Automated CodeRepair for ArkTS](https://arxiv.org/abs/2602.08866) <br>
+[FMBench: Adaptive Large Language Model Output Formatting](https://arxiv.org/abs/2602.06384) <br>
+[DScheLLM: Enabling Dynamic Scheduling through a Fine-Tuned Dual-System Large language Model](https://arxiv.org/abs/2601.09100) <br>
+[Holmes: An Evidence-Grounded LLM Agent for Auditable DDoS Investigation in Cloud Networks](https://arxiv.org/abs/2601.14601) <br>
+[Cochain: Balancing Insufficient and Excessive Collaboration in LLM Agent Workflows](https://arxiv.org/abs/2505.10936) <br>
+[How Few-shot Demonstrations Affect Prompt-based Defenses Against LLM Jailbreak Attacks](https://arxiv.org/abs/2602.04294) <br>
+
+
+
+
+
+**1B**
+
+[Post-Training Quantization of OpenPangu Models for Efficient Deployment on Atlas A2](https://arxiv.org/abs/2512.23367) <br>
+[VisionPangu: A Compact and Fine-Grained Multimodal Assistant with 1.7B Parameters](https://arxiv.org/abs/2603.04957) <br>
+[Neuromem: A Granular Decomposition of the Streaming Lifecycle in External Memory for LLMs](https://arxiv.org/abs/2602.13967) <br>
+[CSSBench: Evaluating the Safety of Lightweight LLMs against Chinese-Specific Adversarial Patterns](https://arxiv.org/abs/2601.00588) <br>
+[A-IO: Adaptive Inference Orchestration for Memory-Bound NPUs](https://arxiv.org/abs/2604.09752) <br>
+[AGZO: Activation-Guided Zeroth-Order Optimization for LLM Fine-Tuning](https://arxiv.org/abs/2601.17261) <br>
+[Combining Adam and its Inverse Counterpart to Enhance Generalization of Deep Learning Optimizers](https://arxiv.org/abs/2603.07122) <br>
+[Characterize LSM-tree Compaction Performance via On-Device LLM Inference](https://arxiv.org/abs/2602.12669v1) <br>
+[PanguMotion: Continuous Driving Motion Forecasting with Pangu Transformers](https://arxiv.org/abs/2603.16196) <br>
+[FMBench: Adaptive Large Language Model Output Formatting](https://arxiv.org/abs/2602.06384) <br>
+[Cochain: Balancing Insufficient and Excessive Collaboration in LLM Agent Workflows](https://arxiv.org/abs/2505.10936) <br>
+[How Few-shot Demonstrations Affect Prompt-based Defenses Against LLM Jailbreak Attacks](https://arxiv.org/abs/2602.04294) <br>
+[Adaptive Confidence Gating in Multi-Agent Collaboration for Efficient and Optimized Code Generation](https://arxiv.org/abs/2601.21469) <br>
+
+
+
+**7B-V1.1**
+
+[SpatialText: A Pure-Text Cognitive Benchmark for Spatial Understanding in Large Language Models](https://arxiv.org/abs/2603.03002) <br>
+[Efficient Reasoning with Balanced Thinking](https://arxiv.org/abs/2603.12372) <br>
+[OSUM-Pangu: An Open-Source Multidimension Speech Understanding Foundation Model Built upon OpenPangu on Ascend NPUs](https://arxiv.org/abs/2603.10862) <br>
+[EvoOpt-LLM: Evolving industrial optimization models with large language models](https://arxiv.org/abs/2602.01082) <br>
+[Accelerating OpenPangu Inference on NPU via Speculative Decoding](https://arxiv.org/abs/2603.03383) <br>
+[CAMERA: Multi-Matrix Joint Compression for MoE Models via Micro-Expert Redundancy Analysis](https://arxiv.org/abs/2508.02322) <br>
+[TypePro: Boosting LLM-Based Type Inference via Inter-Procedural Slicing](https://arxiv.org/abs/2604.02702) <br>
+[UniPruning: Unifying Local Metric and Global Feedback for Scalable Sparse LLMs](https://arxiv.org/abs/2510.03291) <br>
+[Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression](https://arxiv.org/abs/2602.08324) <br>
+[DScheLLM: Enabling Dynamic Scheduling through a Fine-Tuned Dual-System Large language Model](https://arxiv.org/abs/2601.09100) <br>
+[VAR-MATH: Probing True Mathematical Reasoning in LLMS via Symbolic Multi-Instance Benchmarks](https://arxiv.org/abs/2507.12885) <br>
+
+
+**1B-V1.1**
+
+[Silent Inconsistency in Data-Parallel Full Fine-Tuning: Diagnosing Worker-Level Optimization Misalignment](https://arxiv.org/abs/2602.14462) <br>
+[AdapShot: Adaptive Many-Shot In-Context Learning with Semantic-Aware KV Cache Reuse](https://arxiv.org/abs/2605.03644v1) <br>
+[Probe and Skip: Self-Predictive Token Skipping for Efficient Long-Context LLM Inference](https://arxiv.org/abs/2601.13155) <br>
+
+
+
+
+
+
+<details>
+<summary>详细清单</summary>
+  
+[Searching Meta Reasoning Skeleton to Guide LLM Reasoning](https://arxiv.org/abs/2510.04116)
+
+</details>