{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/code/compute-grpo-outcome-advantage","entry":"compute_grpo_outcome_advantage","source":"Syntology graph, per-sample; not an archive number","read_at":"2026-09-24T18:15:14+00:00","claim":"Names are grouped by exact entry-name string. Same-named routines are NOT asserted to be equivalent; 'ran' means executed on a synthesized fixture, not correctness. n_samples_ran = sum of by_status over every status except 'unverified' (ran_draft_wrong and ran_fixture are failures of Syntology's instrument, not of the code); n_papers_ran = papers with at least one such sample.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"},"n_papers":14,"n_papers_ran":2,"units":"n_samples, n_samples_ran, n_samples_fingerprinted and by_status count distinct code bodies (code_sha256); n_places and n_places_pointer_only count places, one per (paper, code body) pair, which is also the unit of the samples list","n_samples":11,"n_samples_ran":2,"n_samples_fingerprinted":2,"n_places":14,"n_places_pointer_only":2,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":2,"ran_fixture":0,"ran":0,"unverified":9},"syntology":{"atlas_url":null,"mcp":null,"mcp_per_sample":{"tool":"get_code","arguments_in":"samples[].mcp_get_code"},"developers":"https://syntology.ai/developers"},"samples":[{"arxiv_id":"2606.10646","paper":"/paper/arxiv-2606-10646","title":"How Does Reasoning Flow? Tracing Attention-Induced Information Flow for Targeted RL in LLMs","date":null,"month_inferred_from_arxiv_id":"2026-06","title_source":"syntology","repo":"Jiayi-Pan/TinyZero","path":"verl/trainer/ppo/core_algos.py","file_url":"https://github.com/Jiayi-Pan/TinyZero/blob/HEAD/verl/trainer/ppo/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"e25af267f6e0b9e8","mcp_get_code":{"code_sha256":"e25af267f6e0b9e8"}},{"arxiv_id":"2606.07705","paper":"/paper/arxiv-2606-07705","title":"SAW: Stage-Aware Dynamic Weighting for Multi-Objective Reinforcement Learning in Large Language Models","date":null,"month_inferred_from_arxiv_id":"2026-06","title_source":"syntology","repo":"Zhaolutuan/SAW","path":"verl/trainer/ppo/core_algos.py","file_url":"https://github.com/Zhaolutuan/SAW/blob/HEAD/verl/trainer/ppo/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"e25af267f6e0b9e8","mcp_get_code":{"code_sha256":"e25af267f6e0b9e8"}},{"arxiv_id":"2605.07465","paper":"/paper/arxiv-2605-07465","title":"SEIF: Self-Evolving Reinforcement Learning for Instruction Following","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"Rainier-rq1/SEIF","path":"verl/trainer/core_algos.py","file_url":"https://github.com/Rainier-rq1/SEIF/blob/HEAD/verl/trainer/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"b7dacadfa17e35d2","mcp_get_code":{"code_sha256":"b7dacadfa17e35d2"}},{"arxiv_id":"2602.15564","paper":"/paper/arxiv-2602-15564","title":"Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQL","date":null,"month_inferred_from_arxiv_id":"2026-02","title_source":"syntology","repo":"Satissss/SquRL","path":"verl/trainer/ppo/core_algos.py","file_url":"https://github.com/Satissss/SquRL/blob/HEAD/verl/trainer/ppo/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"5e44de7d8d0e5d67","mcp_get_code":{"code_sha256":"5e44de7d8d0e5d67"}},{"arxiv_id":"2602.04284","paper":"/paper/arxiv-2602-04284","title":"Agent-Omit: Adaptive Context Omission for Efficient LLM Agents","date":null,"month_inferred_from_arxiv_id":"2026-02","title_source":"syntology","repo":"usail-hkust/Agent-Omit","path":"AgentOmit-RL/verl/agent_trainer/ppo/core_algos.py","file_url":"https://github.com/usail-hkust/Agent-Omit/blob/HEAD/AgentOmit-RL/verl/agent_trainer/ppo/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"9869c756ff538fd1","mcp_get_code":{"code_sha256":"9869c756ff538fd1"}},{"arxiv_id":"2509.26272","paper":"/paper/arxiv-2509-26272","title":"PRPO: Paragraph-level Policy Optimization for Vision-Language Deepfake Detection","date":null,"month_inferred_from_arxiv_id":"2025-09","title_source":"syntology","repo":"tuanrpt/PRPO","path":"prpo/verl/trainer/ppo/core_algos.py","file_url":"https://github.com/tuanrpt/PRPO/blob/HEAD/prpo/verl/trainer/ppo/core_algos.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"a064f095b2208875","mcp_get_code":{"code_sha256":"a064f095b2208875"}},{"arxiv_id":"2509.02522","paper":"/paper/arxiv-2509-02522","title":"Implicit Actor Critic Coupling via a Supervised Learning Framework for RLVR","date":null,"month_inferred_from_arxiv_id":"2025-09","title_source":"syntology","repo":"ritzz-ai/PACS","path":"src/pacs/pacs_core_algos.py","file_url":"https://github.com/ritzz-ai/PACS/blob/HEAD/src/pacs/pacs_core_algos.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"9f098d16adc06043","mcp_get_code":{"code_sha256":"9f098d16adc06043"}},{"arxiv_id":"2507.07966","paper":"/paper/scaling-rl-to-long-videos","title":"Scaling RL to Long Videos","date":"2025-07-10","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"NVlabs/Long-RL","path":"verl/trainer/core_algos.py","file_url":"https://github.com/NVlabs/Long-RL/blob/HEAD/verl/trainer/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"758c7f607efffb52","mcp_get_code":{"code_sha256":"758c7f607efffb52"}},{"arxiv_id":"2507.05386","paper":"/paper/reinforcement-fine-tuning-naturally-mitigates","title":"Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training","date":"2025-07-07","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"zhhvvv/rft_vs_sft","path":"EasyR1/verl/trainer/core_algos.py","file_url":"https://github.com/zhhvvv/rft_vs_sft/blob/HEAD/EasyR1/verl/trainer/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"b7dacadfa17e35d2","mcp_get_code":{"code_sha256":"b7dacadfa17e35d2"}},{"arxiv_id":"2506.01908","paper":"/paper/reinforcement-learning-tuning-for-videollms","title":"Reinforcement Learning Tuning for VideoLLMs: Reward Design and Data Efficiency","date":"2025-06-02","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"appletea233/temporal-r1","path":"verl/trainer/core_algos.py","file_url":"https://github.com/appletea233/temporal-r1/blob/HEAD/verl/trainer/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"17a3ed8bc3471fb5","mcp_get_code":{"code_sha256":"17a3ed8bc3471fb5"}},{"arxiv_id":"2505.16282","paper":"/paper/arpo-end-to-end-policy-optimization-for-gui","title":"ARPO:End-to-End Policy Optimization for GUI Agents with Experience Replay","date":"2025-05-22","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"dvlab-research/arpo","path":"verl/trainer/core_algos.py","file_url":"https://github.com/dvlab-research/arpo/blob/HEAD/verl/trainer/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"b7dacadfa17e35d2","mcp_get_code":{"code_sha256":"b7dacadfa17e35d2"}},{"arxiv_id":"2505.11049","paper":"/paper/2505-11049","title":"GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning","date":"2025-05-16","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"yueliu1999/guardreasoner-vl","path":"train/EasyR1/verl/trainer/core_algos.py","file_url":"https://github.com/yueliu1999/guardreasoner-vl/blob/HEAD/train/EasyR1/verl/trainer/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"dc62e3047c2f6e56","mcp_get_code":{"code_sha256":"dc62e3047c2f6e56"}},{"arxiv_id":"2504.20073","paper":"/paper/ragen-understanding-self-evolution-in-llm","title":"RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning","date":"2025-04-24","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"ragen-ai/ragen","path":"ragen/trainer/core_algos.py","file_url":"https://github.com/ragen-ai/ragen/blob/HEAD/ragen/trainer/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"dd9a34fc6888e4a5","mcp_get_code":{"code_sha256":"dd9a34fc6888e4a5"}},{"arxiv_id":"2504.09696","paper":"/paper/grpo-lead-a-difficulty-aware-reinforcement-1","title":"GRPO-LEAD: A Difficulty-Aware Reinforcement Learning Approach for Concise Mathematical Reasoning in Language Models","date":"2025-04-13","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"aeroplanepaper/GRPO-LEAD","path":"verl/verl/trainer/ppo/core_algos.py","file_url":"https://github.com/aeroplanepaper/GRPO-LEAD/blob/HEAD/verl/verl/trainer/ppo/core_algos.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"b5f62fd7c302812c","mcp_get_code":{"code_sha256":"b5f62fd7c302812c"}}]}