Key papers

• SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning (2026)
  Jinyang Wu et al.
  14.78
• Accurate, Interdisciplinary and Transparent Structure-property Understanding with Deep Native Structural Reasoning (2026)
  Chen Tang et al.
  14.59
• Loop the Loopies! (2026)
  Zitian Gao et al.
  14.05
• Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization (2026)
  Yuxiang Ji et al.
  12.72
• Stale but Stable: Staleness-Adaptive Trust Regions for Stabilizing Asynchronous Reinforcement Learning (2026)
  Junyao Yang et al.
  11.84
• Subliminal Clocks: Latent Time Modelling in Diffusion Language Models (2026)
  Maximo Eduardo Rulli et al.
  11.15
• Group Entropy-Controlled Policy Optimization (2026)
  Guangran Cheng et al.
  10.42
• Understanding Reasoning from Pretraining to Post-Training (2026)
  Jingyan Shen et al.
  10.15
• LLM-as-a-Coach: Experiential Learning for Non-Verifiable Tasks (2026)
  Tianzhu Ye et al.
  9.90
• MultAttnAttrib: Training-Free Multimodal Attribution in Long Document Question Answering (2026)
  Dang Quang Thien Tran et al.
  8.86
• Token-Level Off-Policy Learning for Faithful Generation Under Distribution Shift (2026)
  Zitong Huang et al.
  7.36
• Frames2LoRA: Parametric Video Internalization for Vision-Language Models (2026)
  Manan Suri et al.
  6.69
• SCITUNE: Aligning Large Language Models with Human-Curated Scientific Multimodal Instructions (2023)
  Sameera Horawalavithana et al.
  6.39
• Will Scaling Improve Social Simulation with LLMs? (2026)
  Caleb Ziems et al.
  5.89
• On Improving Faithfulness of Podcasts from Documents (2026)
  Soumya Dutta et al.
  5.89
• Theory-Level Autoformalization: From Isolated Statements to Unified Formal Knowledge Bases (2026)
  Marcus J. Min et al.
  5.88
• Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES (2026)
  Hunter Heidenreich
  5.87
• Train the Model, Not the Reader: Decodability Supervision for Verifiable Activation Explanations (2026)
  Hiskias Dingeto
  5.85
• WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search (2026)
  Xiaoshuai Song et al.
  5.49
• Do You Need a Frontier Model as a Citation Verifier? Benchmarking Rubric LLMs for Deep-Research Source Attribution (2026)
  Ethan Leung et al.
  5.49
• The Hard Decision Layer: Evidence for Committed Inference in Transformers (2026)
  Ashwath Vaithinathan Aravindan et al.
  5.49
• Opaque Epistemic Mediation: How LLM Deployment Configurations Shape the Validation of Pseudo-Science (2026)
  Davide Scarso et al.
  5.49
• Multi-Objective Exploration and Preference Optimization via Mutual Information (2026)
  Hongyan Xie et al.
  5.01
• From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations (2026)
  Shiyi Ling et al.
  5.01
• ICDAR 2026 HIPE-OCRepair Competition on LLM-Assisted OCR Post-Correction for Historical Documents (2026)
  Maud Ehrmann et al.
  5.01
• CityBehavEx: A Scalable and Empirically Validated LLM-Assisted Urban Simulation Platform (2026)
  Gustavo H. Santos et al.
  5.01
• LLM-Driven AutoML for Cross-Lingual Handwritten OCR: Closed-Loop Neural Architecture Search with GPT-5, GPT-4o, and Claude Sonnet 4 (2026)
  Mobina Kashaniyan et al.
  5.01
• Measuring Reward-Seeking via Contrastive Belief Updates (2026)
  Axel H{\o}jmark et al.
  5.01
• TopoGuard: Graph Theory Based Defenses Against Split-Knowledge Attacks on RAG (2026)
  Chahana Dahal et al.
  5.01
• The Storyteller in the Model: Narrative Pattern Inheritance, Escalation Dynamics, and Alignment Governance in LLMs (2026)
  Adam Rigby et al.
  5.01
• The Geometry of Personality: Activation Steering with Jungian Cognitive Functions (2026)
  Liu Zai (University of Glasgow) et al.
  5.01
• A Comparative Evaluation of Embeddings and LLMs in a Greek Book Publisher Setting - The CUP Dataset (2026)
  Katerina Papantoniou et al.
  5.01
• Evaluation design conditions the expert-vs-auto MeSH gap: a controlled comparison of bag-of-words and BiomedBERT on the Cohen benchmark (2026)
  Samuel M. Okoe-Mensah
  5.01
• Learning What Matters: Supervising Sparse Attention Routing with Causal Evidence Sets (2026)
  Jim Allchin
  5.01
• Data Quality over Capacity: Internalizing Documents into LoRA Adapters for Closed-Book QA (2026)
  Joan Figuerola Hurtado
  5.01
• J-CoT: Chain-of-Thought in J-Space (2026)
  Junde Wu et al.
  5.01
• DWT-Fusion: A Signal-Based Framework for Training-Free LLM-Generated Text Detection (2026)
  Mehmet Batuhan \"Ozda\c{s} et al.
  5.01
• Enough is as good as a feast: A Comprehensive Analysis of How Reinforcement Learning Mitigates Task Conflicts in LLMs (2026)
  Zixuan Ren et al.
  5.01
• FSE: Continual Learning for Named Entity Recognition by Fast-Slow Experts (2026)
  Yunan Zhang et al.
  5.01
• MEUSLI: a Multilingual Projector for LLM-based ASR and Beyond (2026)
  Lorenzo Concina et al.
  5.01
• Dynamic Commonsense Coordination for Empathetic Response Generation (2026)
  Zhengyu Qi
  5.01
• A Factorial Study of Synthetic Data Generation for Low-Resource Machine Translation using Grammar Books (2026)
  Varun Ghat Ravikumar et al.
  5.01
• Multimodal Knowledge Edit-Scoped Generalization for Online Recursive MLLM Editing (2026)
  Siyuan Li et al.
  4.39
• When Does In-Context Search Help? A Sampling-Complexity Theory of Reflection-Driven Reasoning (2026)
  Yotam Wolf et al.
  4.39
• Dissociating the Internal Representations of Sycophancy in LLMs (2026)
  Anthony Baez et al.
  4.39
• Diagnosing and Calibrating Tool-Call Boundary Drift in Multi-Teacher On-Policy Distillation (2026)
  Jiabin Shen et al.
  4.39
• Understanding Interpretation Difficulty in Harmful Online Communication: Insights from Cybercrime Communities (2026)
  Tomohiro Okatsu et al.
  4.39
• A Word-Level Digital Reader of the Prasthanatrayi with Sankara's Bhasya: Corpus, Method, and an Open, Offline Reading Aid for the Advaita Vedanta Canon (2026)
  Tamal Maharaj
  4.39
• R^3: Advertisement Compliance Rectification via Group-Relative Experience Extractor and Curriculum Reinforcement (2026)
  Yuan Chen et al.
  4.39
• TF-Engram: A Train-Free Engram with SSD-Backed Memory for Large Language Models (2026)
  Yutang Ma et al.
  4.39
• DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting (2026)
  Hong-Kai Zheng et al.
  4.39
• SynthAVE: Scalable Synthetic Labeling for E-Commerce with LLM-Arena Validation (2026)
  Andrea Scarinci et al.
  4.39
• Beyond Attack-Success Rate: Action-Graded Severity Scale for Tool-Using AI Agents (2026)
  Harry Owiredu-Ashley
  4.39
• Interpreting Latent CoT Reasoning as Dynamical Systems (2026)
  Sabari Iyyappan Duraipandian et al.
  4.39
• Silent Failures in Quantized LLM Reasoning: A Taxonomy-Based Analysis of Hollow Convergence and Failure Mode Shifts (2026)
  Renuka Oladri et al.
  4.39
• Polarization Detection: A Hybrid Approach with AfroXLMR-Social and DeBERTa for Low- and High-Resource Settings (2026)
  Muhammad Abdullahi Said
  4.39
• A Stepwise Questioning Expert-Editor Multi-Agent Framework for Long-Document Summarization (2026)
  Lingyun Shen et al.
  4.39
• Demographic Prompting at Scale: When More Attributes Hurt LLM--Human Agreement (2026)
  Mahammed Kamruzzaman et al.
  4.39
• Non-binary bottom-up constituency parsing without arity actions (2026)
  Jungyeul Park et al.
  4.39
• Abstractiveness Metrics for Evaluating Text Summarization: A Refined Formulation with Empirical Validation (2026)
  Praveenkumar Katwe et al.
  4.39