How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift | Paper Citation Record

Observation 15c20bf2-6b42-4332-97ca-365fa69fed23 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Training Verifiers to Solve Math Word Problems

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:08.300365Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b9ed988-c8cd-455e-86de-0ffd23a790bc · outbound

This paper cites A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:08.456270Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4db81838-fe98-44f2-b42a-6cf38c5448e7 · outbound

This paper cites Shashwat Goel, Rishi Hazra, Dulhan Jayalath, Timon Willi, Parag Jain, William F Shen, Ilias Leontiadis, Francesco Barbieri, Yoram Bachrach, Jonas Geiping, et al.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Shashwat Goel, Rishi Hazra, Dulhan Jayalath, Timon Willi, Parag Jain, William F Shen, Ilias Leontiadis, Francesco Barbieri, Yoram Bachrach, Jonas Geiping, et al

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:08.575041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1d13e2b-96d8-4e02-8bde-7704aebf7b2f · outbound

This paper cites Towards an AI co-scientist.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Towards an AI co-scientist

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:08.709873Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 893a902e-5c71-448f-b6ab-b9601a943054 · outbound

This paper cites The Llama 3 Herd of Models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift The Llama 3 Herd of Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:08.843992Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f1c8b897-c2c1-455b-bbcd-af557fd70f26 · outbound

This paper cites Alignment faking in large language models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Alignment faking in large language models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:08.964328Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b686f350-5d9c-45fd-b691-3b5b48c10978 · outbound

This paper cites Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.041384Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 67bbd316-71c9-4e0c-aa08-37a9e75ccdb4 · outbound

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

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.145063Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a95bba1c-e718-4851-bb67-d551f67aa950 · outbound

This paper cites Val-bench: Measuring value alignment in language models.arXiv preprint arXiv:2510.05465,.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Val-bench: Measuring value alignment in language models.arXiv preprint arXiv:2510.05465,

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.221267Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 60817586-e035-49c2-9647-8334b689781a · outbound

This paper cites What is in Your Safe Data? Identifying Benign Data that Breaks Safety.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift What is in Your Safe Data? Identifying Benign Data that Breaks Safety

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.310328Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 93edbb92-ec61-45e5-a8d3-97889d1de0cf · outbound

This paper cites Understanding catastrophic forgetting in language models via implicit inference.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Understanding catastrophic forgetting in language models via implicit inference

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.409877Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed4e70e1-67fb-424a-937f-79b556eadb57 · outbound

This paper cites LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.563432Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cd3fdea8-e9d5-448b-8413-30ca16bb3482 · outbound

This paper cites Holistic Evaluation of Language Models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Holistic Evaluation of Language Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.803411Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fa646ea2-3a36-4082-bb39-13ac93b65480 · outbound

This paper cites TruthfulQA: Measuring How Models Mimic Human Falsehoods.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift TruthfulQA: Measuring How Models Mimic Human Falsehoods

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.929971Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f1373da-6470-4aa8-8236-455916a01d57 · outbound

This paper cites Natural emergent misalignment from reward hacking in production RL.arXiv preprint arXiv:2511.18397,.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Natural emergent misalignment from reward hacking in production RL.arXiv preprint arXiv:2511.18397,

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.066489Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ac092efd-0bf9-41b7-8ef4-d2f87a8cea56 · outbound

This paper cites HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.182971Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da97eac8-1fb3-4ae9-8c6b-0d23d3ed6c0e · outbound

This paper cites CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.447865Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18ae449e-e1b3-4d26-bef0-54597ca2b9c6 · outbound

This paper cites Steering Llama 2 via Contrastive Activation Addition.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Steering Llama 2 via Contrastive Activation Addition

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.618734Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a7d4085-942d-4811-a70d-16f853adca04 · outbound

This paper cites BBQ: A Hand-Built Bias Benchmark for Question Answering.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift BBQ: A Hand-Built Bias Benchmark for Question Answering

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.724281Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e611791f-350a-4d55-914d-eda50aacda53 · outbound

This paper cites Evaluating Frontier Models for Stealth and Situational Awareness.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Evaluating Frontier Models for Stealth and Situational Awareness

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.876626Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a909cb6d-3143-4ced-a1df-19e32ac9708d · outbound

This paper cites Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.978883Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2d90204a-e197-4ddb-829f-a2dc575a1c93 · outbound

This paper cites Proximal Policy Optimization Algorithms.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Proximal Policy Optimization Algorithms

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.097152Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation faa928a4-7c31-4015-a6e7-8f38383d3c1d · outbound

This paper cites Towards understanding sycophancy in language models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Towards understanding sycophancy in language models

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.291461Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9799853d-1213-49ec-9fed-4a72e5596838 · outbound

This paper cites SEAT: Sparse Entity-Aware Tuning for Knowledge Adaptation while Preserving Epistemic Abstention.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift SEAT: Sparse Entity-Aware Tuning for Knowledge Adaptation while Preserving Epistemic Abstention

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.373496Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ecc72661-25ad-4704-bebd-5a06674b1d0d · outbound

This paper cites Rethinking rubric generation for improving llm judge and reward modeling for open-ended tasks.arXiv preprint arXiv:2602.05125, 2026b.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Rethinking rubric generation for improving llm judge and reward modeling for open-ended tasks.arXiv preprint arXiv:2602.05125, 2026b

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.472483Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f4383b94-2b9f-4d14-9b4a-b5ffc5bd0505 · outbound

This paper cites Efficiency vs.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Efficiency vs

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.574118Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a576681-4e36-4855-a189-130589f46210 · outbound

This paper cites Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.683306Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ccb2f31-6291-4be4-81d2-34523d9259e8 · outbound

This paper cites Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.904578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44ce3f33-e5bb-444d-b68f-21fc76dd0456 · outbound

This paper cites Continual Learning for Large Language Models: A Survey.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Continual Learning for Large Language Models: A Survey

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.986628Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c758f0d3-8683-4048-894a-f186754125ba · outbound

This paper cites Qwen2.5 Technical Report.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Qwen2.5 Technical Report

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.128064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3282e5e5-ac55-46e1-aca2-a48f7d1b8956 · outbound

This paper cites Qwen3 Technical Report.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Qwen3 Technical Report

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.204085Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6dc0aa6b-640a-4bd9-8fbc-554c13c04373 · outbound

This paper cites Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.284016Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0bc464dc-398e-45a6-a3c0-388d06036cc7 · outbound

This paper cites Removing RLHF Protections in GPT-4 via Fine-Tuning.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Removing RLHF Protections in GPT-4 via Fine-Tuning

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.338931Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation af3ebd5a-7c02-4ae3-8603-1ef177ca39df · outbound

This paper cites Instruction-Following Evaluation for Large Language Models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Instruction-Following Evaluation for Large Language Models

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.407447Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 67ea0e6d-4bc8-479c-8011-a47dc483ee69 · outbound

This paper cites The path not taken: RLVR provably learns off the principals.arXiv preprint arXiv:2511.08567,.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift The path not taken: RLVR provably learns off the principals.arXiv preprint arXiv:2511.08567,

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.507333Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dd97763e-1c09-4b9f-824d-79d1451e595e · outbound

This paper cites Representation Engineering: A Top-Down Approach to AI Transparency.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Representation Engineering: A Top-Down Approach to AI Transparency

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.594577Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0bb0f0b0-23d6-40c2-a88b-c47c2bfe9f17 · outbound

This paper cites 15 A.2 KL-Regularized SFT.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift 15 A.2 KL-Regularized SFT

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.714401Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29f6243c-f362-4ae0-8bd9-ab00a9b7110f · outbound

This paper cites GRPO-based RLVR is trained to a fixed number of steps with almost all reaching reward saturation, while SFT and KL-SFT use the fixed epoch budgets in Table.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift GRPO-based RLVR is trained to a fixed number of steps with almost all reaching reward saturation, while SFT and KL-SFT use the fixed epoch budgets in Table

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.794413Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21c96e9d-c54a-45a6-873c-e0280e23026d · outbound

This paper cites an unresolved cited work.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Unresolved cited work

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.861870Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 240f51fc-e227-4227-86ff-10f2289e7486 · outbound

This paper cites We report the headline metric, the preferred direction, and the aggregation procedure used when benchmarks contain multiple subtasks.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift We report the headline metric, the preferred direction, and the aggregation procedure used when benchmarks contain multiple subtasks

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:12.908742Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9305019-ea83-44a8-a90a-13cdd5fe1d29 · outbound

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

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.200447Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 47ed0b9f-43c9-4d4a-b30e-33b65240793b · outbound

This paper cites an unresolved cited work.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Unresolved cited work

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:10.281592Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 53feb5b4-32ca-43f4-b1cf-38789ac341f3 · outbound

This paper cites Breaking the safety-capability tradeoff: Reinforcement learning with verifiable rewards maintains safety guardrails in LLMs.arXiv preprint arXiv:2511.21050,.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Breaking the safety-capability tradeoff: Reinforcement learning with verifiable rewards maintains safety guardrails in LLMs.arXiv preprint arXiv:2511.21050,

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:08.124547Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2f170256-bcb4-45cb-b0a5-042892823d5f · outbound

This paper cites TACO: Topics in Algorithmic COde generation dataset.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift TACO: Topics in Algorithmic COde generation dataset

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.658983Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 715f6f3d-a357-440f-8bbd-1efd2c690acf · outbound

This paper cites Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:09.499800Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c2b3e20b-dbc5-4ab0-8f0f-77258d1ed2fb · outbound

This paper cites Program Synthesis with Large Language Models.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Program Synthesis with Large Language Models

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:07.870753Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be94e655-cd29-4999-9bef-8b9fb268711a · outbound

This paper cites Evaluating Large Language Models Trained on Code.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Evaluating Large Language Models Trained on Code

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:07.946955Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca54fa19-a4f1-410c-986d-48769cb5dcee · outbound

This paper cites Persona features control emergent misalignment.arXiv preprint arXiv:2506.19823,.

How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift Persona features control emergent misalignment.arXiv preprint arXiv:2506.19823,

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-08-02T07:44:11.838709Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers