Key papers

• MemoryLLM: Plug-n-Play Interpretable Feed-Forward Memory for Transformers (2026)
  Ajay Jaiswal et al.
  1.94
• Stop the Flip-Flop: Context-Preserving Verification for Fast Revocable Diffusion Decoding (2026)
  Yanzheng Xiang et al.
  1.94
• On-Policy Self-Distillation for Reasoning Compression (2026)
  Hejian Sang et al.
  1.94
• Thinking in Uncertainty: Mitigating Hallucinations in MLRMs with Latent Entropy-Aware Decoding (2026)
  Zhongxing Xu et al.
  1.94
• SeGPruner: Semantic-Geometric Visual Token Pruner for 3D Question Answering (2026)
  Wenli Li et al.
  1.94
• Where Should Diffusion Enter a Language Model? Geometry-Guided Hidden-State Replacement (2026)
  Injin Kong et al.
  1.94
• Not All Disagreement Is Learnable: Token Teachability in On-Policy Distillation (2026)
  Yuanyi Wang et al.
  1.94
• Optical Reasoning: Rethinking Images as an Expressive Reasoning Medium Beyond Text (2026)
  Yutong Bian et al.
  1.94
• APPO: Agentic Procedural Policy Optimization (2026)
  Xucong Wang et al.
  1.94
• Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models (2026)
  Cheng-Yu Yang et al.
  1.94
• Grouped Query Experts: Mixture-of-Experts on GQA Self-Attention (2026)
  Vishesh Tripathi et al.
  1.94
• Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints (2026)
  Fangzheng Li et al.
  1.94
• Stop When Reasoning Converges: Semantic-Preserving Early Exit for Reasoning Models (2026)
  Dehai Min et al.
  1.89
• Dynin-Omni: Omnimodal Unified Large Diffusion Language Model (2026)
  Jaeik Kim et al.
  1.83
• UniMixer: A Unified Architecture for Scaling Laws in Recommendation Systems (2026)
  Mingming Ha et al.
  1.83
• Small Vision-Language Models are Smart Compressors for Long Video Understanding (2026)
  Junjie Fei et al.
  1.83
• InfiniteScienceGym: An Unbounded, Procedurally-Generated Benchmark for Scientific Analysis (2026)
  Oliver Bentham et al.
  1.83
• TIP: Token Importance in On-Policy Distillation (2026)
  Yuanda Xu et al.
  1.83
• A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression (2026)
  Jincheng Ren et al.
  1.83
• CurveStream: Boosting Streaming Video Understanding in MLLMs via Curvature-Aware Hierarchical Visual Memory Management (2026)
  Chao Wang et al.
  1.78
• BEAVER: A Training-Free Hierarchical Prompt Compression Method via Structure-Aware Page Selection (2026)
  Zhengpei Hu et al.
  1.78
• Model Context Protocol (MCP) Tool Descriptions Are Smelly! Towards Improving AI Agent Efficiency with Augmented MCP Tool Descriptions (2026)
  Mohammed Mehedi Hasan et al.
  1.72
• AgentConductor: Topology Evolution for Multi-Agent Competition-Level Code Generation (2026)
  Siyu Wang et al.
  1.72
• DUET-VLM: Dual stage Unified Efficient Token reduction for VLM Training and Inference (2026)
  Aditya Kumar Singh et al.
  1.72
• O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning (2025)
  Haotian Luo et al.
  1.28
• Cramming 1568 Tokens into a Single Vector and Back Again: Exploring the Limits of Embedding Space Capacity (2025)
  Yuri Kuratov et al.
  1.28
• Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation (2025)
  Sucheng Ren et al.
  1.28
• Teaching Metric Distance to Autoregressive Multimodal Foundational Models (2025)
  Jiwan Chung et al.
  1.28
• LLMVoX: Autoregressive Streaming Text-to-Speech Model for Any LLM (2025)
  Sambal Shikhar et al.
  1.28
• QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension (2025)
  Yongdong Luo et al.
  1.28
• Communication-Efficient Language Model Training Scales Reliably and Robustly: Scaling Laws for DiLoCo (2025)
  Zachary Charles et al.
  1.28
• Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers (2025)
  Weiming Ren et al.
  1.28
• Fixing 7,400 Bugs for 1$: Cheap Crash-Site Program Repair (2025)
  Han Zheng et al.
  1.28
• Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space (2025)
  Zhen Zhang et al.
  1.28
• QwenLong-CPRS: Towards infty-LLMs with Dynamic Context Optimization (2025)
  Weizhou Shen et al.
  1.28
• ARM: Adaptive Reasoning Model (2025)
  Siye Wu et al.
  1.28
• Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment (2025)
  Xiaojun Jia et al.
  1.28
• DINGO: Constrained Inference for Diffusion LLMs (2025)
  Tarun Suresh et al.
  1.28
• D-AR: Diffusion via Autoregressive Models (2025)
  Ziteng Gao et al.
  1.28
• SiLVR: A Simple Language-based Video Reasoning Framework (2025)
  Ce Zhang et al.
  1.28
• Improving large language models with concept-aware fine-tuning (2025)
  Michael K. Chen et al.
  1.28
• SRFT: A Single-Stage Method with Supervised and Reinforcement Fine-Tuning for Reasoning (2025)
  Yuqian Fu et al.
  1.28
• StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context Modeling (2025)
  Meng Wei et al.
  1.28
• A Practical Two-Stage Recipe for Mathematical LLMs: Maximizing Accuracy with SFT and Efficiency with Reinforcement Learning (2025)
  Hiroshi Yoshihara et al.
  1.28
• VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning (2025)
  Senqiao Yang et al.
  1.28
• On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification (2025)
  Yongliang Wu et al.
  1.28
• Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning (2025)
  Lijie Yang et al.
  1.28
• DeCRED: Decoder-Centric Regularization for Encoder-Decoder Based Speech Recognition (2025)
  Alexander Polok et al.
  1.28
• Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning (2025)
  Vaishnavi Shrivastava et al.
  1.28
• Lavida-O: Elastic Large Masked Diffusion Models for Unified Multimodal Understanding and Generation (2025)
  Shufan Li et al.
  1.28
• Thinking-Free Policy Initialization Makes Distilled Reasoning Models More Effective and Efficient Reasoners (2025)
  Xin Xu et al.
  1.28
• No Tokens Wasted: Leveraging Long Context in Biomedical Vision-Language Models (2025)
  Min Woo Sun et al.
  1.28
• Training-Free Group Relative Policy Optimization (2025)
  Yuzheng Cai et al.
  1.28
• HyperAgent: Leveraging Hypergraphs for Topology Optimization in Multi-Agent Communication (2025)
  Heng Zhang et al.
  1.28
• Distilled Decoding 2: One-step Sampling of Image Auto-regressive Models with Conditional Score Distillation (2025)
  Enshu Liu et al.
  1.28
• Mitigating Attention Sinks and Massive Activations in Audio-Visual Speech Recognition with LLMS (2025)
  Anand et al.
  1.28
• MMPersuade: A Dataset and Evaluation Framework for Multimodal Persuasion (2025)
  Haoyi Qiu et al.
  1.28
• The Sequential Edge: Inverse-Entropy Voting Beats Parallel Self-Consistency at Matched Compute (2025)
  Aman Sharma et al.
  1.28
• Can Visual Input Be Compressed? A Visual Token Compression Benchmark for Large Multimodal Models (2025)
  Tianfan Peng et al.
  1.28
• Cross-Lingual Supervision improves Large Language Models Pre-training (2023)
  Andrea Schioppa et al.
  —