# Accumulating Context Changes the Beliefs of Language Models

arXiv: https://arxiv.org/abs/2511.01805 · HTML: https://mlsanity.com/paper/2511.01805/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Builds on 63 of its 108 references that are in the map.
- First posted on arXiv 3 Nov 2025.

## Builds on (most cited first, 30 of 63)

- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Jailbreaking Black Box Large Language Models in Twenty Queries](https://mlsanity.com/paper/2310.08419/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models](https://mlsanity.com/paper/2310.04451/) (2023)
- [Tree of Attacks: Jailbreaking Black-Box LLMs Automatically](https://mlsanity.com/paper/2312.02119/) (2023)
- [The Internal State of an LLM Knows When It's Lying](https://mlsanity.com/paper/2304.13734/) (2023)
- [BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents](https://mlsanity.com/paper/2504.12516/) (2025)
- [How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs](https://mlsanity.com/paper/2401.06373/) (2024)
- [Reasoning Models Don't Always Say What They Think](https://mlsanity.com/paper/2505.05410/) (2025)
- [Using cognitive psychology to understand GPT-3](https://mlsanity.com/paper/2206.14576/) (2022)
- [Agent Workflow Memory](https://mlsanity.com/paper/2409.07429/) (2024)
- [Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers](https://mlsanity.com/paper/2409.04109/) (2024)
- [WebThinker: Empowering Large Reasoning Models with Deep Research Capability](https://mlsanity.com/paper/2504.21776/) (2025)
- [DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads](https://mlsanity.com/paper/2410.10819/) (2024)
- [Debating with More Persuasive LLMs Leads to More Truthful Answers](https://mlsanity.com/paper/2402.06782/) (2024)
- [MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues](https://mlsanity.com/paper/2402.14762/) (2024)
- [MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent](https://mlsanity.com/paper/2507.02259/) (2025)
- [WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models](https://mlsanity.com/paper/2406.18510/) (2024)
- [Evaluating the Moral Beliefs Encoded in LLMs](https://mlsanity.com/paper/2307.14324/) (2023)
- [SkillWeaver: Web Agents can Self-Improve by Discovering and Honing Skills](https://mlsanity.com/paper/2504.07079/) (2025)
- [Do LLMs exhibit human-like response biases? A case study in survey design](https://mlsanity.com/paper/2311.04076/) (2023)
- [LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet](https://mlsanity.com/paper/2408.15221/) (2024)
- [ReviewerGPT? An Exploratory Study on Using Large Language Models for Paper Reviewing](https://mlsanity.com/paper/2306.00622/) (2023)
- [Inducing Programmatic Skills for Agentic Tasks](https://mlsanity.com/paper/2504.06821/) (2025)
- [Agentic Reasoning: A Streamlined Framework for Enhancing LLM Reasoning with Agentic Tools](https://mlsanity.com/paper/2502.04644/) (2025)
- [MT-Eval: A Multi-Turn Capabilities Evaluation Benchmark for Large Language Models](https://mlsanity.com/paper/2401.16745/) (2024)
- [The Ideation-Execution Gap: Execution Outcomes of LLM-Generated versus Human Research Ideas](https://mlsanity.com/paper/2506.20803/) (2025)

## Built on by (most cited first, 28 of 28)

- [Characterizing Delusional Spirals through Human-LLM Chat Logs](https://mlsanity.com/paper/2603.16567/) (2026)
- [On Safety Risks in Experience-Driven Self-Evolving Agents](https://mlsanity.com/paper/2604.16968/) (2026)
- [BeliefShift: Benchmarking Temporal Belief Consistency and Opinion Drift in LLM Agents](https://mlsanity.com/paper/2603.23848/) (2026)
- [Belief Memory: Agent Memory Under Partial Observability](https://mlsanity.com/paper/2605.05583/) (2026)
- [Linear representations in language models can change dramatically over a conversation](https://mlsanity.com/paper/2601.20834/) (2026)
- [Position: Modular Memory is the Key to Continual Learning Agents](https://mlsanity.com/paper/2603.01761/) (2026)
- [BayesBench: Evaluating LLM Belief Trajectories Under Multi-Turn Evidence Accumulation](https://mlsanity.com/paper/2606.30850/) (2026)
- [Recalling Too Well: Sycophancy Evaluation and Mitigation in Memory-Augmented Models](https://mlsanity.com/paper/2606.10949/) (2026)
- [Ads in AI Chatbots? An Analysis of How Large Language Models Navigate Conflicts of Interest](https://mlsanity.com/paper/2604.08525/) (2026)
- [Useless but Safe? Benchmarking Utility Recovery with User Intent Clarification in Multi-Turn Conversations](https://mlsanity.com/paper/2604.27093/) (2026)
- [Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space](https://mlsanity.com/paper/2605.12412/) (2026)
- [Illusions of Confidence? Diagnosing LLM Truthfulness via Neighborhood Consistency](https://mlsanity.com/paper/2601.05905/) (2026)
- [Intersectional Sycophancy: How Perceived User Demographics Shape False Validation in Large Language Models](https://mlsanity.com/paper/2604.11609/) (2026)
- [Topics as Proxies for Sociodemographics: How Conversational Context Affects LLM Answers](https://mlsanity.com/paper/2606.02776/) (2026)
- [Beyond Sycophancy: Structured Resistance and Compliance in LLM Moral Reasoning](https://mlsanity.com/paper/2607.21558/) (2026)
- [Persuasion Propagation: Does Persuasion Change AI Agent Behavior?](https://mlsanity.com/paper/2602.00851/) (2026)
- [To Nuke or Not to Nuke: LLMs' (Missing) Ethical Reasoning and Actions in a High-Stakes Decision-Making Simulation](https://mlsanity.com/paper/2606.08310/) (2026)
- [Evidence-Informed LLM Beliefs for Continual Scientific Discovery](https://mlsanity.com/paper/2606.29182/) (2026)
- [What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates](https://mlsanity.com/paper/2607.02507/) (2026)
- [Can LLMs Take Retrieved Information with a Grain of Salt?](https://mlsanity.com/paper/2605.06919/) (2026)
- [Relational Positioning as a Measurable Risk Object: History-Carried Lock-in and Self-Confabulation in Multi-Turn Human-AI Dialogue](https://mlsanity.com/paper/2607.11437/) (2026)
- [DelusionEval: Measuring Delusion-Linked Behaviors in AI Chatbots](https://mlsanity.com/paper/2608.05004/) (2026)
- [Identifying Harm in Personalized, Generative AI Systems Requires User-Centered Auditing at the Interaction Level](https://mlsanity.com/paper/2608.14692/) (2026)
- [Belief Cascades Drive Persuasion in LLM Agent Networks](https://mlsanity.com/paper/2608.25152/) (2026)
- [Do LLMs Change Their Minds Like Humans? Diagnosing Human--LLM Divergence in Single-Turn Persuasion Judgments](https://mlsanity.com/paper/2608.29803/) (2026)
- [Illusory Truth or Mere Exposure? Model-Dependent Repetition Effects in LLM-Based Social Media Simulations](https://mlsanity.com/paper/2609.36278/) (2026)
- [EurekaBench: Measuring Agentic Ability to Discover New Scientific Insights](https://mlsanity.com/paper/2610.00492/) (2026)
- [Not All Answers Are Contextually Persuadable: Inference Dynamics in Large Language Models under Contextual Influence](https://mlsanity.com/paper/2610.04791/) (2026)

## Abstract (arXiv)

Language model (LM) assistants are increasingly used in applications such as brainstorming and research. Improvements in memory and context size have allowed these models to become more autonomous, which has also resulted in more text accumulation in their context windows without explicit user intervention. This comes with a latent risk: the belief profiles of models -- their understanding of the world as manifested in their responses or actions -- may silently change as context accumulates. This can lead to subtly inconsistent user experiences, or shifts in behavior that deviate from the original alignment of the models. In this paper, we explore how accumulating context by engaging in interactions and processing text -- talking and reading -- can change the beliefs of language models, as manifested in their responses and behaviors. Our results reveal that models' belief profiles are highly malleable: GPT-5 exhibits a 54.7% shift in its stated beliefs after 10 rounds of discussion about moral dilemmas and queries about safety, while Grok 4 shows a 27.2% shift on political issues after reading texts from the opposing position. We also examine models' behavioral changes by designing tasks that require tool use, where each tool selection corresponds to an implicit belief. We find that these changes align with stated belief shifts, suggesting that belief shifts will be reflected in actual behavior in agentic systems. Our analysis exposes the hidden risk of belief shift as models undergo extended sessions of talking or reading, rendering their opinions and actions unreliable.