Reasoning Fine-Tuning Induces Persistent Latent Policy States | Paper Citation Record

Observation 33a8c4d3-bf7e-42a6-b333-a605efb0fb5e · outbound

This paper cites 2023 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2023 , eprint=

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:13.575026Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0237123c-6c4a-4e5f-adee-38562c24232e · outbound

This paper cites ICML 2025 Workshop on Reliable and Responsible Foundation Models , year=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States ICML 2025 Workshop on Reliable and Responsible Foundation Models , year=

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:13.663724Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cba94a23-644f-4f6c-9b5f-3a7de9259228 · outbound

This paper cites Proceedings of the 41st International Conference on Machine Learning , articleno =.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Proceedings of the 41st International Conference on Machine Learning , articleno =

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:13.791427Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 825cca83-bfc2-474e-8259-d4b1cd4b9007 · outbound

This paper cites 2026 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2026 , eprint=

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.005522Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9cae750f-2587-4ca1-88f3-e75583e151c8 · outbound

This paper cites The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.134693Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6728a5c1-0b1f-4253-97c4-046d8771266b · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.242377Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8650d4b9-d380-4d2a-89a0-f46024c1e7d1 · outbound

This paper cites On the Limits of.

Reasoning Fine-Tuning Induces Persistent Latent Policy States On the Limits of

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.327297Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 467d64c5-f2cd-4572-a170-e50cd354927b · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.406132Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 042f060e-c17d-44f4-939f-c7e173964bee · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.471044Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b27331d8-64f4-47b5-b859-f09b6d073b63 · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.598063Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 59874cf3-8d33-4391-937f-60ac25e2127c · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.677547Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab57c9b8-672b-4e80-8a1e-80b946677828 · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.747523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3ea5a647-9542-4900-a049-97d160b676a9 · outbound

This paper cites an unresolved cited work.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Unresolved cited work

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:14.815561Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 65c73470-88b7-444d-8548-0bc050766d3f · outbound

This paper cites and Spragins, John D.

Reasoning Fine-Tuning Induces Persistent Latent Policy States and Spragins, John D

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.019700Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 481ad52a-d756-478c-8517-25952180d99c · outbound

This paper cites On the Direction of.

Reasoning Fine-Tuning Induces Persistent Latent Policy States On the Direction of

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.094412Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 57f14358-b1f0-4f17-9a2d-a3fce4a535b9 · outbound

This paper cites 2026 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2026 , eprint=

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.262132Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a4ddcdd1-beb6-48fe-8f41-19a0e287696b · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.356343Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1ce70e93-0578-4921-87ec-31d8686c4f5e · outbound

This paper cites and Fu, K.

Reasoning Fine-Tuning Induces Persistent Latent Policy States and Fu, K

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.450612Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ee69454c-69e7-4093-9127-01f9b8345d97 · outbound

This paper cites Neural computation , volume=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Neural computation , volume=

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.533323Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4395fd2e-2538-4e89-838d-91a570e3e027 · outbound

This paper cites 2017 , editor =.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2017 , editor =

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.610158Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cd2e82d8-aea0-439d-ab6b-df7d719fcfff · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.678014Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 54220faa-9214-465a-8618-85660cb82660 · outbound

This paper cites The Fourteenth International Conference on Learning Representations , year=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States The Fourteenth International Conference on Learning Representations , year=

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.801597Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2b926bd8-6e17-461e-bdab-6e899cd33315 · outbound

This paper cites Mechanistic Interpretability Workshop at NeurIPS 2025 , year=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Mechanistic Interpretability Workshop at NeurIPS 2025 , year=

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:15.980313Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 630c407c-a66f-4480-b977-196daf945a96 · outbound

This paper cites Towards Understanding Fine-Tuning Mechanisms of.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Towards Understanding Fine-Tuning Mechanisms of

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.039869Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9e5e1ac0-cb93-4dab-8106-9df9f2059402 · outbound

This paper cites ICLR 2025 Workshop on Building Trust in Language Models and Applications , year=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States ICLR 2025 Workshop on Building Trust in Language Models and Applications , year=

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.144526Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d296e457-83f4-4039-a39f-d774b220f5b7 · outbound

This paper cites Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.228601Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 50574a29-ae63-4fe4-bb1e-ffc014cf2002 · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.343248Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 54b48bc6-6e46-434f-89c6-f4365975738d · outbound

This paper cites Workshop on Reasoning and Planning for Large Language Models , year=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Workshop on Reasoning and Planning for Large Language Models , year=

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.413360Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da6c3141-549f-43aa-8ee2-91ea6bfdbea9 · outbound

This paper cites Measuring Chain of Thought Faithfulness by Unlearning Reasoning Steps.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Measuring Chain of Thought Faithfulness by Unlearning Reasoning Steps

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.520411Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6fa5c60-3bb7-4993-83f7-8ddb32e3a250 · outbound

This paper cites 2021 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2021 , eprint=

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.643332Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f8116245-94a6-411a-bc09-21ba5fa65f8c · outbound

This paper cites 2024 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2024 , eprint=

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.866739Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 53c5c773-b076-460e-994b-3049f8fac1f0 · outbound

This paper cites 2024 , url =.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2024 , url =

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:16.981690Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 61f7b793-bbd8-4996-afd4-85a667aad898 · outbound

This paper cites QwQ-32B: Embracing the Power of Reinforcement Learning , url =.

Reasoning Fine-Tuning Induces Persistent Latent Policy States QwQ-32B: Embracing the Power of Reinforcement Learning , url =

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:17.452678Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9d0feb81-2050-4136-93cb-a57b02161405 · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:17.707886Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f5990075-718f-479f-9710-f153f3c414d7 · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:17.816565Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fee474c1-646b-46b4-b2c0-a875375e7dd2 · outbound

This paper cites 2025 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , eprint=

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:17.937789Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f8cd739-d4a3-4d91-80f4-3e502332c454 · outbound

This paper cites 2024 , eprint=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2024 , eprint=

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.051644Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c1a7d186-5f2b-4244-86a1-8d38051610ce · outbound

This paper cites 2024 , month = aug, journal =.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2024 , month = aug, journal =

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.171431Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 225151fe-8573-4ab5-b0f2-7e6505fbcb51 · outbound

This paper cites 2026 , url=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2026 , url=

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.289118Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 428c6fe0-8b92-49d3-8936-030695d31b1f · outbound

This paper cites 2025 , url=.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 2025 , url=

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.387352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7bb7bf73-2cc3-4f78-9efc-07ade3fe236f · outbound

This paper cites Ackerson and K.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Ackerson and K

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.507344Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e88f86d1-e470-4804-b859-fb76df477f91 · outbound

This paper cites Llama 3 model card.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Llama 3 model card

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.625152Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44f25d29-4d44-4d87-b9ae-14a3052e1a72 · outbound

This paper cites Qwen Technical Report.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Qwen Technical Report

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.727168Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3cbf0cc6-f2e8-4005-85a6-b52cc9a20e8a · outbound

This paper cites On the identifiability of switching dynamical systems.

Reasoning Fine-Tuning Induces Persistent Latent Policy States On the identifiability of switching dynamical systems

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:18.872972Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5218849b-2c28-4c8b-b8ee-efb2367ce13e · outbound

This paper cites Bogdan, Uzay Macar, Neel Nanda, and Arthur Conmy.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Bogdan, Uzay Macar, Neel Nanda, and Arthur Conmy

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.017904Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20f4bf1c-0488-4917-9586-2376a3cff0d4 · outbound

This paper cites A statistical physics of language model reasoning.

Reasoning Fine-Tuning Induces Persistent Latent Policy States A statistical physics of language model reasoning

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.135142Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 512e6320-024c-4caa-8445-8b1cf574c0ab · outbound

This paper cites SEAL : Steerable reasoning calibration of large language models for free.

Reasoning Fine-Tuning Induces Persistent Latent Policy States SEAL : Steerable reasoning calibration of large language models for free

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.233415Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ace043ea-d3d4-43c4-88c9-7942c4c9783b · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Training Verifiers to Solve Math Word Problems

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.284845Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6b08b06a-9f2c-4e64-af04-aab8727a9034 · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

Reasoning Fine-Tuning Induces Persistent Latent Policy States DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.344259Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 70284a98-88aa-4380-a7e5-48c4bca322a4 · outbound

This paper cites Asymptotic properties of the maximum likelihood estimator in autoregressive models with markov regime.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Asymptotic properties of the maximum likelihood estimator in autoregressive models with markov regime

Reference 60

Resolution: verified exact; doi, observed 2026-08-01T15:18:36.075659Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 77f71155-c3a6-4da7-8e08-3e46fc698976 · outbound

This paper cites Variational learning for switching state-space models.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Variational learning for switching state-space models

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.513576Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5eaa8c9a-8439-4081-ae5c-a8f6bb238749 · outbound

This paper cites Measuring Mathematical Problem Solving With the MATH Dataset.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Measuring Mathematical Problem Solving With the MATH Dataset

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.587543Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3210d02a-0435-42ce-ac1e-c12e06c530c9 · outbound

This paper cites On the direction of RLVR updates for LLM reasoning: Identification and exploitation.

Reasoning Fine-Tuning Induces Persistent Latent Policy States On the direction of RLVR updates for LLM reasoning: Identification and exploitation

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.658340Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 097ff543-1f93-46fe-aa8e-85ee20435285 · outbound

This paper cites Tulu 3: Pushing Frontiers in Open Language Model Post-Training.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Tulu 3: Pushing Frontiers in Open Language Model Post-Training

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.738268Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2f6276b5-af19-497b-ba48-1973f23e5736 · outbound

This paper cites The CoT Encyclopedia: Analyzing, Predicting, and Controlling how a Reasoning Model will Think.

Reasoning Fine-Tuning Induces Persistent Latent Policy States The CoT Encyclopedia: Analyzing, Predicting, and Controlling how a Reasoning Model will Think

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.811207Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f94be62e-2ada-4fe9-8fb0-fe73221fb720 · outbound

This paper cites Clue: Non-parametric verification from experience via hidden-state clustering.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Clue: Non-parametric verification from experience via hidden-state clustering

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.877258Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 148555f6-d956-4d39-8031-a158ce7cf927 · outbound

This paper cites Bayesian Learning and Inference in Recurrent Switching Linear Dynamical Systems.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Bayesian Learning and Inference in Recurrent Switching Linear Dynamical Systems

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:19.948636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9a20706a-5886-4b1f-98de-690212d3e067 · outbound

This paper cites Learning a generative meta-model of llm activations, 2026.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Learning a generative meta-model of llm activations, 2026

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.033854Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f5477018-f50b-47dc-8d13-76beba001d38 · outbound

This paper cites Thought Branches: Interpreting LLM Reasoning Requires Resampling.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Thought Branches: Interpreting LLM Reasoning Requires Resampling

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.107692Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb1d2b8b-99c5-40e1-8dfc-91c1afe523c8 · outbound

This paper cites Narrow finetuning leaves clearly readable traces in the activation differences.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Narrow finetuning leaves clearly readable traces in the activation differences

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.156420Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7fada9f-7740-45c3-aa0a-411da92ede04 · outbound

This paper cites Diab, Virginia Smith, and Dawn Song.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Diab, Virginia Smith, and Dawn Song

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.213159Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05a6d9c7-35b3-44c4-89bc-836597cd8b16 · outbound

This paper cites La, Duy M.

Reasoning Fine-Tuning Induces Persistent Latent Policy States La, Duy M

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.267179Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1029cfcf-3514-4aa6-a2ee-6900768ac5b8 · outbound

This paper cites an unresolved cited work.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Unresolved cited work

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.324657Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 105d9ba2-7d4e-4233-a7dc-a4bf246b3867 · outbound

This paper cites Demystifying reasoning dynamics with mutual information: Thinking tokens are information peaks in LLM reasoning.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Demystifying reasoning dynamics with mutual information: Thinking tokens are information peaks in LLM reasoning

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.400811Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ef28362d-1aac-4f39-9ed9-9a994a4888a0 · outbound

This paper cites Learnable latent embeddings for joint behavioural and neural analysis.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Learnable latent embeddings for joint behavioural and neural analysis

Reference 75

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.504062Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5990ec5-5c43-40ac-a081-b8e8e931a8ec · outbound

This paper cites DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.

Reasoning Fine-Tuning Induces Persistent Latent Policy States DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.617290Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation abc2bec8-ab5a-4f28-be14-fd6ae8b393ff · outbound

This paper cites Understanding reasoning in thinking language models via steering vectors.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Understanding reasoning in thinking language models via steering vectors

Reference 77

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.713420Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation de19a86b-b3f8-46dc-a0c7-27e2c4d16e13 · outbound

This paper cites Base Models Know How to Reason, Thinking Models Learn When.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Base Models Know How to Reason, Thinking Models Learn When

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.782109Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 68b545c0-b727-40d3-8cf8-685a9d3bdbde · outbound

This paper cites Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.893289Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 088cf89d-45ac-41fc-b511-ae95e7b496a1 · outbound

This paper cites Towards understanding fine-tuning mechanisms of LLM s via circuit analysis.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Towards understanding fine-tuning mechanisms of LLM s via circuit analysis

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:20.964020Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ef769a69-4967-46e1-9560-0f82b6e00b25 · outbound

This paper cites MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark.

Reasoning Fine-Tuning Induces Persistent Latent Policy States MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.070905Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1b1a916-9964-45b0-91f3-c4a1b61dcfc8 · outbound

This paper cites Reasoning-Finetuning Repurposes Latent Representations in Base Models.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Reasoning-Finetuning Repurposes Latent Representations in Base Models

Reference 82

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.171395Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a095c327-b401-4bce-ad08-19f547317a1d · outbound

This paper cites Rank-1 loras encode interpretable reasoning signals.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Rank-1 loras encode interpretable reasoning signals

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.241977Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 757871bd-ffcf-445d-95be-d910dad14927 · outbound

This paper cites On the limits of RLVR : Support, entropy, and the illusion of reasoning.

Reasoning Fine-Tuning Induces Persistent Latent Policy States On the limits of RLVR : Support, entropy, and the illusion of reasoning

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.323304Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0548a8f8-b974-4f50-87f4-108e828d447e · outbound

This paper cites CTRLS : Chain-of-thought reasoning via latent state transition.

Reasoning Fine-Tuning Induces Persistent Latent Policy States CTRLS : Chain-of-thought reasoning via latent state transition

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.396068Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e79d1087-996e-49e7-b07e-f4815b7116d5 · outbound

This paper cites Yakowitz and John D.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Yakowitz and John D

Reference 86

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.468535Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11e2a532-030f-4bcb-959e-fd2c96be1b83 · outbound

This paper cites Qwen2 Technical Report.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Qwen2 Technical Report

Reference 87

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.564657Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8d4e677c-f82b-45cf-8808-8cf8d0623749 · outbound

This paper cites Qwen2.5 Technical Report.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Qwen2.5 Technical Report

Reference 88

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.608057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14968147-d138-4ad7-9bf1-d7be20e534cb · outbound

This paper cites Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?.

Reasoning Fine-Tuning Induces Persistent Latent Policy States Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?

Reference 89

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.697846Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e8206db-ed4d-4f84-bf78-8b65574e8cf3 · outbound

This paper cites 1.4 Million Open-Source Distilled Reasoning Dataset to Empower Large Language Model Training.

Reasoning Fine-Tuning Induces Persistent Latent Policy States 1.4 Million Open-Source Distilled Reasoning Dataset to Empower Large Language Model Training

Reference 90

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.796399Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0555707-067a-4fbd-964d-8e4fd93601b9 · outbound

This paper cites The path not taken: Rlvr provably learns off the principals.

Reasoning Fine-Tuning Induces Persistent Latent Policy States The path not taken: Rlvr provably learns off the principals

Reference 91

Resolution: unresolved; no resolver link, observed 2026-08-01T15:16:21.881919Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers