{"arxiv_id":"2607.28048","incoming":{"papers":0,"sentences":0,"groups":[],"unlabelled_sentences":0},"outgoing":{"papers":19,"sentences":31,"groups":[{"kind":"data_used","label":"Uses data or values from","papers":2,"sentences":2,"quotes":[{"arxiv_id":"1704.05179","title":"SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine","url":"/paper/1704.05179","section":"Experiments","sentence":"The main experiment reports each benchmark's native hard-success or exact-match score on the test partitions across SearchQA [ref] , SpreadsheetBench [ref] , DocVQA [ref] , LiveMath [ref] , and ALFWorld [ref] .","intent_p":0.89,"contests_p":null},{"arxiv_id":"2604.01754","title":"LiveMathematicianBench: A Live Benchmark for Research-Level Mathematical Reasoning with Proof Sketches","url":"/paper/2604.01754","section":"Experiments","sentence":"The main experiment reports each benchmark's native hard-success or exact-match score on the test partitions across SearchQA [ref] , SpreadsheetBench [ref] , DocVQA [ref] , LiveMath [ref] , and ALFWorld [ref] .","intent_p":0.89,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":13,"sentences":20,"quotes":[{"arxiv_id":"2606.26790","title":"OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning","url":"/paper/2606.26790","section":"Knowledge Distillation for LLM Agents","sentence":"Recent work uses skills or privileged signals mainly as training-time supervision for policy learning or reinforcement learning [ref] .","intent_p":0.97,"contests_p":null},{"arxiv_id":"2602.08234","title":"SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning","url":"/paper/2602.08234","section":"Knowledge Distillation for LLM Agents","sentence":"Recent work uses skills or privileged signals mainly as training-time supervision for policy learning or reinforcement learning [ref] .","intent_p":0.97,"contests_p":null},{"arxiv_id":"2604.03128","title":"Self-Distilled RLVR","url":"/paper/2604.03128","section":"Knowledge Distillation for LLM Agents","sentence":"Recent work uses skills or privileged signals mainly as training-time supervision for policy learning or reinforcement learning [ref] .","intent_p":0.97,"contests_p":null},{"arxiv_id":"2604.10674","title":"Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents","url":"/paper/2604.10674","section":"Knowledge Distillation for LLM Agents","sentence":"Recent work uses skills or privileged signals mainly as training-time supervision for policy learning or reinforcement learning [ref] .","intent_p":0.97,"contests_p":null},{"arxiv_id":"2506.14728","title":"AgentDistill: Training-Free Agent Distillation with Generalizable MCP Boxes","url":"/paper/2506.14728","section":"Knowledge Distillation for LLM Agents","sentence":"In agent settings, prior work distills tool-using agent behavior through trajectory-based training [ref] , while AgentDistill transfers reusable external modules without training [ref] .","intent_p":0.96,"contests_p":null}]}],"unlabelled_sentences":9},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/2607.28048"}}