{"total":52,"rows":[{"id":"2502.12143","title":"Small Models Struggle to Learn from Strong Reasoners","created":"2025-02-17","cited_by_count":113,"area":"lr"},{"id":"2505.11711","title":"Reinforcement Learning Finetunes Small Subnetworks in Large Language Models","created":"2025-05-16","cited_by_count":49,"area":null},{"id":"2503.12759","title":"RAG-RL: Advancing Retrieval-Augmented Generation via RL and Curriculum Learning","created":"2025-03-17","cited_by_count":30,"area":"rag"},{"id":"2510.01624","title":"Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead","created":"2025-10-02","cited_by_count":28,"area":"lr"},{"id":"2510.00526","title":"Beyond Log Likelihood: Probability-Based Objectives for Supervised Fine-Tuning across the Model Capability Continuum","created":"2025-10-01","cited_by_count":18,"area":null},{"id":"2502.16182","title":"IPO: Your Language Model is Secretly a Preference Classifier","created":"2025-02-22","cited_by_count":17,"area":"alignment"},{"id":"2601.14249","title":"Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment","created":"2026-01-20","cited_by_count":13,"area":"lr"},{"id":"2506.06964","title":"Offline RL by Reward-Weighted Fine-Tuning for Conversation Optimization","created":"2025-06-08","cited_by_count":10,"area":"alignment"},{"id":"2510.03988","title":"The Signal is in the Steps: Local Scoring for Reasoning Data Selection","created":"2025-10-05","cited_by_count":8,"area":"distill"},{"id":"2605.19433","title":"Backtracking When It Strays: Mitigating Dual Exposure Biases in LLM Reasoning Distillation","created":"2026-05-19","cited_by_count":8,"area":"lr"},{"id":"2510.13291","title":"Higher Satisfaction, Lower Cost: A Technical Report on How LLMs Revolutionize Meituan's Intelligent Interaction Systems","created":"2025-10-15","cited_by_count":7,"area":null},{"id":"2512.20908","title":"Where Did This Sentence Come From? Tracing Provenance in LLM Reasoning Distillation","created":"2025-12-24","cited_by_count":6,"area":"lr"},{"id":"2604.07941","title":"Large Language Model Post-Training: A Unified View of Off-Policy and On-Policy Learning","created":"2026-04-09","cited_by_count":6,"area":null},{"id":"2511.02833","title":"In Good GRACEs: Principled Teacher Selection for Knowledge Distillation","created":"2025-11-04","cited_by_count":5,"area":"distill"},{"id":"2606.04261","title":"Can Generalist Agents Automate Data Curation?","created":"2026-06-02","cited_by_count":5,"area":null},{"id":"2508.04329","title":"Forgetting: A New Mechanism Towards Better Large Language Model Fine-tuning","created":"2025-08-06","cited_by_count":4,"area":null},{"id":"2604.07892","title":"Data Selection for Multi-turn Dialogue Instruction Tuning","created":"2026-04-09","cited_by_count":4,"area":null},{"id":"2505.05327","title":"RICo: Refined In-Context Contribution for Automatic Instruction-Tuning Data Selection","created":"2025-05-08","cited_by_count":3,"area":null},{"id":"2505.23765","title":"From Chat Logs to Collective Insights: Aggregative Question Answering","created":"2025-05-29","cited_by_count":3,"area":null},{"id":"2512.04555","title":"ADAPT: Learning Task Mixtures for Budget-Constrained Instruction Tuning","created":"2025-12-04","cited_by_count":3,"area":null},{"id":"2507.17107","title":"Reinforcement Learning Fine-Tunes a Sparse Subnetwork in Large Language Models","created":"2025-07-23","cited_by_count":2,"area":"alignment"},{"id":"2602.15257","title":"How to Train Your Long-Context Visual Document Model","created":"2026-02-16","cited_by_count":2,"area":"mllm"},{"id":"2605.10415","title":"Aligning LLM Uncertainty with Human Disagreement in Subjectivity Analysis","created":"2026-05-11","cited_by_count":2,"area":null},{"id":"2606.12360","title":"Anatomy of Post-Training: Using Interpretability to Characterize Data and Shape the Learning Signal","created":"2026-06-10","cited_by_count":2,"area":null},{"id":"2603.13790","title":"Aligning the Query Space: Greedy Information Projection for Language Model Data Selection","created":"2026-03-14","cited_by_count":1,"area":null},{"id":"2605.19285","title":"Are Rationales Necessary and Sufficient? Tuning LLMs for Explainable Misinformation Detection","created":"2026-05-19","cited_by_count":1,"area":null},{"id":"2605.26872","title":"When the Strongest Teacher Is Not the Best Teacher: Student-Centric Answer Selection","created":"2026-05-26","cited_by_count":1,"area":"lr"},{"id":"2605.30651","title":"LARK: Learnability-Grounded Trajectory Selection for Efficient Reasoning Distillation","created":"2026-05-28","cited_by_count":1,"area":"lr"},{"id":"2608.13721","title":"Capacity-Dependent Effects of Data Selection for Reasoning","created":"2026-08-13","cited_by_count":1,"area":null},{"id":"2506.23979","title":"TaP: A Taxonomy-Guided Framework for Automated and Scalable Preference Data Generation","created":"2025-06-30","cited_by_count":0,"area":null},{"id":"2509.06463","title":"Accelerate Scaling of LLM Finetuning via Quantifying the Coverage and Depth of Instruction Set","created":"2025-09-08","cited_by_count":0,"area":null},{"id":"2510.05534","title":"Revisiting Self-Play Preference Optimization: On the Role of Prompt Difficulty","created":"2025-10-07","cited_by_count":0,"area":"alignment"},{"id":"2602.11220","title":"Patch the Distribution Mismatch: RL Rewriting Agent for Stable Off-Policy SFT","created":"2026-02-11","cited_by_count":0,"area":"agents"},{"id":"2604.03557","title":"When Do Hallucinations Arise? A Graph Perspective on the Evolution of Path Reuse and Path Compression","created":"2026-04-04","cited_by_count":0,"area":"lr"},{"id":"2604.12748","title":"Generating Effective CoT Traces for Mitigating Causal Hallucination","created":"2026-04-14","cited_by_count":0,"area":"lr"},{"id":"2604.13538","title":"Synthesizing Instruction-Tuning Datasets with Contrastive Decoding","created":"2026-04-15","cited_by_count":0,"area":"synth"},{"id":"2604.25132","title":"What Makes Good Instruction-Tuning Data? An In-Context Learning Perspective","created":"2026-04-28","cited_by_count":0,"area":null},{"id":"2605.09404","title":"Let the Target Select for Itself: Data Selection via Target-Aligned Paths","created":"2026-05-10","cited_by_count":0,"area":null},{"id":"2605.12906","title":"Data Difficulty and the Generalization--Extrapolation Tradeoff in LLM Fine-Tuning","created":"2026-05-13","cited_by_count":0,"area":null},{"id":"2605.15012","title":"Boosting Reinforcement Learning with Verifiable Rewards via Randomly Selected Few-Shot Guidance","created":"2026-05-14","cited_by_count":0,"area":"lr"},{"id":"2605.29288","title":"Diagnosing Harmful Continuation in Answer-Correct Long-CoT Training Traces","created":"2026-05-28","cited_by_count":0,"area":"lr"},{"id":"2605.30857","title":"MADS: Model-Aware Diverse Core Set Selection for Instruction Tuning","created":"2026-05-29","cited_by_count":0,"area":null},{"id":"2606.11189","title":"A Unifying Lens on Supervised Fine-Tuning Through Target Distribution Design","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2607.03166","title":"KARMA: Knowledge graph-based Automated Reasoning Materialization and Alignment","created":"2026-07-03","cited_by_count":0,"area":"lr"},{"id":"2608.16926","title":"Data-DPO: Direct Preference Optimization for Target Model Data Selection in LLM Post-Training","created":"2026-08-05","cited_by_count":0,"area":"alignment"},{"id":"2608.17268","title":"Understanding Curriculum Learning in Large Language Models via Cross-Difficulty Optimization Dynamics","created":"2026-08-18","cited_by_count":0,"area":null},{"id":"2608.29118","title":"Emergent Misalignment Is Not Magical","created":"2026-08-29","cited_by_count":0,"area":null},{"id":"2609.33780","title":"Selecting Diverse SFT Traces Improves Post-RL Generalization","created":"2026-09-27","cited_by_count":0,"area":"lr"},{"id":"2609.36246","title":"Learning from Teacher Continuations at Student States","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2610.00949","title":"PG-SFT: Balancing Capability Acquisition and Retention in Offline Agent Fine-Tuning","created":"2026-10-01","cited_by_count":0,"area":"agents"},{"id":"2610.05831","title":"Selecting Long-Horizon Trajectories for Reliable and Efficient Terminal-Agent Training","created":"2026-10-05","cited_by_count":0,"area":null},{"id":"2610.18001","title":"Match the Canvas, Not Just the Answer: Predictability-Aware SFT for Diffusion Language Models","created":"2026-10-13","cited_by_count":0,"area":"diffusion"}],"next_cursor":null}