{"total":415,"papers":[{"id":"arxiv:2606.29150","title":"Flow Reasoning Models: Turning Flows Into Efficient Recurrent Reasoners","date":"2026-06-28","section":"Reasoning and the \"physics\" of language models","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/8be1801f0f562711039ced2b42f324434c6ab829f544b8d736bc85405b64021b/","source_url":"https://arxiv.org/abs/2606.29150","text_available":true},{"id":"arxiv:2306.02451","title":"For SALE: State-Action Representation Learning for Deep Reinforcement Learning","date":"2023-06-04","section":"Reinforcement learning","quality":0.048804,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/952e0e96fff2857b76e6b62a24d57ca4f6ac86ad8e8a21cb4d31a4dc051f8885/","source_url":"https://arxiv.org/abs/2306.02451","text_available":true},{"id":"arxiv:2509.25123","title":"From $f(x)$ and $g(x)$ to $f(g(x))$: LLMs Learn New Skills in RL by Composing Old Ones","date":"2025-09-29","section":"Post-training","quality":-0.057993,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/7fdfe7abe26ee324fe7abad3ce802148163129a7c157d77bdec859c990868e03/","source_url":"https://arxiv.org/abs/2509.25123","text_available":true},{"id":"arxiv:2405.14838","title":"From Explicit CoT to Implicit CoT: Learning to Internalize CoT Step by Step","date":"2024-05-23","section":"Reasoning and the \"physics\" of language models","quality":-0.083004,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a242b6148f6f84466ce4315ac4937278586a242dd3f847146e142f3c7e27c39c/","source_url":"https://arxiv.org/abs/2405.14838","text_available":true},{"id":"arxiv:2605.22074","title":"From Reasoning Chains to Verifiable Subproblems: Curriculum Reinforcement Learning Enables Credit Assignment for LLM Reasoning","date":"2026-05-21","section":"Post-training","quality":0.202613,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/e51c03db071e17a4d5403433f1037e79c3cda795e0b20f6a490df35167411dd0/","source_url":"https://arxiv.org/abs/2605.22074","text_available":true},{"id":"arxiv:2505.17117","title":"From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning","date":"2025-05-21","section":"NeuroAI","quality":0.424878,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/e0837eb0f0d0e479142b8ae6d2ef144168e5e16c610ad22479076c9d606be084/","source_url":"https://arxiv.org/abs/2505.17117","text_available":true},{"id":"arxiv:2309.10232","title":"Fully parallel optical matrix-matrix multiplication","date":"2023-09-19","section":"Other","quality":-0.758186,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/5462c0d76ea10b3a4230fec0672d24778a1a010ba26bddcda2f5feb0875996fd/","source_url":"https://arxiv.org/abs/2309.10232","text_available":true},{"id":"arxiv:2605.19376","title":"Generative Recursive Reasoning","date":"2026-05-19","section":"Reasoning and the \"physics\" of language models","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/795bad9bb1b1a063090fbb925ba6e68bf30211de0ecffd75d56f858b28321ae7/","source_url":"https://arxiv.org/abs/2605.19376","text_available":true},{"id":"arxiv:2507.19457","title":"GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning","date":"2025-07-25","section":"Post-training","quality":0.450458,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/c1ef72d96fa3292590a6a3d112ab509840081b59aaa771cf42b8b0e8ef62ba08/","source_url":"https://arxiv.org/abs/2507.19457","text_available":true},{"id":"arxiv:2002.05202","title":"GLU Variants Improve Transformer","date":"2020-02-12","section":"LLMs: architectures, context, training","quality":-0.585411,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/e052a2974593d29355ab45c301766ac6f5545ac1bc81d605e78d3adc71e9e2e4/","source_url":"https://arxiv.org/abs/2002.05202","text_available":true},{"id":"arxiv:2602.18037","title":"Gradient Regularization Mitigates Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards","date":"2026-02-20","section":"Post-training","quality":0.140712,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/75161700a0f3e03055116a289c82e9f9fa7d8fee7a6f802e116216eac3a70e31/","source_url":"https://arxiv.org/abs/2602.18037","text_available":true},{"id":"arxiv:2405.20233","title":"Grokfast: Accelerated Grokking by Amplifying Slow Gradients","date":"2024-05-30","section":"Data, training, optimization","quality":0.183974,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/b62d50c44fbd3554702d96f717a5ee25029a28ceed9346ab27ff64d1d08ba07e/","source_url":"https://arxiv.org/abs/2405.20233","text_available":true},{"id":"arxiv:2405.15071","title":"Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization","date":"2024-05-23","section":"Reasoning and the \"physics\" of language models","quality":0.461893,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/1cd7fa83eca0b480c3bf9e2490c043b70e937777252a40b196ae7801a0620141/","source_url":"https://arxiv.org/abs/2405.15071","text_available":true},{"id":"openreview:hcQfTsVnBo","title":"Grokking Group Multiplication with Cosets","date":"2023-12-11","section":"Reasoning and the \"physics\" of language models","quality":0.433766,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/daa23510ca0765cf7cb802f4ab922f167481195d3b31bd012b5e4298470c342d/","source_url":"https://openreview.net/forum?id=hcQfTsVnBo","text_available":true},{"id":"arxiv:2507.18071","title":"Group Sequence Policy Optimization","date":"2025-07-24","section":"Post-training","quality":0.050742,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/69430db904efc7978803673806cc9338004ffd8cf92199f3349dfbfa8a44e440/","source_url":"https://arxiv.org/abs/2507.18071","text_available":true},{"id":"arxiv:2505.10978","title":"Group-in-Group Policy Optimization for LLM Agent Training","date":"2025-05-16","section":"Post-training","quality":0.31133,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/2fa466eef15ebc0489217004e5ebb80efd37db9675a603331ccc3b7da055d147/","source_url":"https://arxiv.org/abs/2505.10978","text_available":true},{"id":"url:cameronrwolfe.substack.com/p/grpo-tricks","title":"GRPO++: Tricks for Making RL Actually Work (Cameron Wolfe)","date":"2026-01-05","section":"Post-training","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d170bc3da5b8e6a055fd511719fcc3764635236bac638947bd0a018412245eb5/","source_url":"https://cameronrwolfe.substack.com/p/grpo-tricks","text_available":false},{"id":"arxiv:2604.20659","title":"GRPO-VPS: Enhancing Group Relative Policy Optimization with Verifiable Process Supervision for Effective Reasoning","date":"2026-04-22","section":"Post-training","quality":-0.004395,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/42caf0f9c992f5df8bac4d0f7e45c3a5bf57879661f41febd01a047694d52628/","source_url":"https://arxiv.org/abs/2604.20659","text_available":true},{"id":"arxiv:2405.16684","title":"gzip Predicts Data-dependent Scaling Laws","date":"2024-05-26","section":"Data, training, optimization","quality":-0.037016,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/8d5456f36b61f2ce9a45da0889b5f0318dc816cab90ee57d54bba06246041de3/","source_url":"https://arxiv.org/abs/2405.16684","text_available":true},{"id":"arxiv:2410.04444","title":"Gödel Agent: A Self-Referential Agent Framework for Recursive Self-Improvement","date":"2024-10-06","section":"Agents, open-endedness, AGI","quality":-0.050839,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/d04ae5d37f10547be79c728efed48a629e631ab5384155564e52558e12ff4362/","source_url":"https://arxiv.org/abs/2410.04444","text_available":true},{"id":"url:openai.com/index/harness-engineering","title":"Harness Engineering (OpenAI)","date":"2026-08-25","section":"Harness","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a64858f28b8f0c5dd1c3a4da212807c69004781880993ef54e5c383d4c9efbe5/","source_url":"https://openai.com/index/harness-engineering/","text_available":false},{"id":"arxiv:2609.01437","title":"HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?","date":"2026-09-01","section":"Harness","quality":-0.042613,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/0e91a157b5aa6d15aa804a3849a31a55661a73d6c1470ff918d72163f5fe2076/","source_url":"https://arxiv.org/abs/2609.01437","text_available":true},{"id":"url:lilianweng.github.io/posts/2026-07-04-harness","title":"Harnesses for models (Lilian Weng, 2026)","date":"2026-07-04","section":"Harness","quality":null,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/c1ea249334ba128a9fff468e25fa4f8d13d4cccd13661ee130bec020ebb6463b/","source_url":"https://lilianweng.github.io/posts/2026-07-04-harness/","text_available":false},{"id":"arxiv:2605.13821","title":"Harnessing Agentic Evolution","date":"2026-05-13","section":"Agents, open-endedness, AGI","quality":0.05538,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/1f1060a507e2bf439d2934d610827d1b3177d85c90c1a2cab483608a56de5138/","source_url":"https://arxiv.org/abs/2605.13821","text_available":true}],"next":"https://vinnik-dmitry07.github.io/favorite-papers/agents/catalog-0-7.html"}