Cognition-of-Thought Elicits Social-Aligned Reasoning in Large Language Models

• URL: http://arxiv.org/abs/2509.23441v2
• Date: Tue, 14 Oct 2025 15:38:48 GMT
• Title: Cognition-of-Thought Elicits Social-Aligned Reasoning in Large Language Models
• Authors: Xuanming Zhang, Yuxuan Chen, Samuel Yeh, Sharon Li,
• Abstract summary: Large language models (LLMs) excel at complex reasoning but can still exhibit harmful behaviors.<n>This paper introduces Cognition-of-Thought (CooT), a novel decoding-time framework that equips LLMs with an explicit cognitive self-monitoring loop.
• Score: 17.381122321801556
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Large language models (LLMs) excel at complex reasoning but can still exhibit harmful behaviors. Current alignment strategies typically embed safety into model weights, making these controls implicit, static, and difficult to modify. This paper introduces Cognition-of-Thought (CooT), a novel decoding-time framework that equips LLMs with an explicit cognitive self-monitoring loop. CooT couples a standard text Generator with a cognitive Perceiver that continuously monitors the unfolding sequence. The Perceiver uses a structured, precedence-based hierarchy of principles (e.g., safety over obedience) to detect potential misalignments as they arise. When violations are flagged, CooT intervenes by rolling back the generation to the point of error and regenerating under injected guidance that combines universal social priors with context-specific warnings. CooT thus transforms alignment from a fixed property into an explicit, dynamic, and auditable process active during inference, allowing for flexible policy updates without retraining the model. Extensive experiments across multiple benchmarks and model families confirm that CooT consistently improves safety and social reasoning performance.

Related papers

• The Emergence of Lab-Driven Alignment Signatures: A Psychometric Framework for Auditing Latent Bias and Compounding Risk in Generative AI [0.0]
  This paper introduces a novel auditing framework to quantify latent trait estimation under ordinal uncertainty.<n>The research audits nine leading models across dimensions including Optimization Bias, Sycophancy, and Status-Quo Legitimization.
  arXiv  Detail & Related papers  (2026-02-19T06:56:01Z)
• Character as a Latent Variable in Large Language Models: A Mechanistic Account of Emergent Misalignment and Conditional Safety Failures [70.48661957773449]
  Emergent Misalignment refers to a failure mode in which fine-tuning large language models on narrowly scoped data induces broadly misaligned behavior.<n>Across multiple domains and model families, we find that fine-tuning models on data exhibiting specific character-level dispositions induces substantially stronger and more transferable misalignment than incorrect-advice fine-tuning.
  arXiv  Detail & Related papers  (2026-01-30T15:28:42Z)
• GAVEL: Towards rule-based safety through activation monitoring [2.337566423505956]
  Large language models (LLMs) are increasingly paired with activation-based monitoring to detect and prevent harmful behaviors.<n>Existing activation safety approaches, trained on broad misuse datasets, struggle with poor precision, limited flexibility, and lack of interpretability.<n>This paper introduces a new paradigm: rule-based activation safety, inspired by rule-sharing practices in cybersecurity.
  arXiv  Detail & Related papers  (2026-01-27T16:31:39Z)
• CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphs [53.199517625701475]
  CoG is a training-free framework inspired by Dual-Process Theory that mimics the interplay between intuition and deliberation.<n>CoG significantly outperforms state-of-the-art approaches in both accuracy and efficiency.
  arXiv  Detail & Related papers  (2026-01-16T07:27:40Z)
• Cracking IoT Security: Can LLMs Outsmart Static Analysis Tools? [1.8549313085249322]
  This work presents the first comprehensive evaluation of Large Language Models (LLMs) across a multi-category interaction threat taxonomy.<n>We benchmark Llama 3.1 8B, Llama 70B, GPT-4o, Gemini-2.5-Pro, and DeepSeek-R1 across zero-, one-, and two-shot settings.<n>Our findings show that while LLMs exhibit promising semantic understanding, their accuracy degrades significantly for threats requiring cross-rule structural reasoning.
  arXiv  Detail & Related papers  (2026-01-02T04:17:36Z)
• TopoReformer: Mitigating Adversarial Attacks Using Topological Purification in OCR Models [2.8272661103123258]
  Adversarially perturbed images of text can cause OCR systems to produce misleading or incorrect transcriptions from seemingly invisible changes to humans.<n>Existing defenses, such as adversarial training, input preprocessing, or post-recognition correction, are often model-specific, computationally expensive, and affect performance on unperturbed inputs.<n>TopoReformer is introduced, a model-agnostic reformation pipeline that mitigates adversarial perturbations while preserving the structural integrity of text images.
  arXiv  Detail & Related papers  (2025-11-19T19:01:51Z)
• Large Reasoning Models Learn Better Alignment from Flawed Thinking [56.08883934423522]
  Large reasoning models (LRMs) "think" by generating structured chain-of-thought (CoT) before producing a final answer.<n>We propose RECAP, a principled reinforcement learning (RL) method for post-training that explicitly teaches models to override flawed reasoning trajectories.
  arXiv  Detail & Related papers  (2025-10-01T14:15:43Z)
• AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models [62.70575022567081]
  We propose AdvChain, an alignment paradigm that teaches models dynamic self-correction through adversarial CoT tuning.<n>Our work establishes a new direction for building more robust and reliable reasoning models.
  arXiv  Detail & Related papers  (2025-09-29T04:27:23Z)
• LatentGuard: Controllable Latent Steering for Robust Refusal of Attacks and Reliable Response Generation [4.29885665563186]
  LATENTGUARD is a framework that combines behavioral alignment with supervised latent space control for interpretable and precise safety steering.<n>Our results show significant improvements in both safety controllability and response interpretability without compromising utility.
  arXiv  Detail & Related papers  (2025-09-24T07:31:54Z)
• Rule Encoding and Compliance in Large Language Models: An Information-Theoretic Analysis [0.0]
  The design of safety-critical agents based on large language models (LLMs) requires more than simple prompt engineering.<n>This paper presents a comprehensive information-theoretic analysis of how rule encodings in system prompts influence attention mechanisms and compliance behaviour.
  arXiv  Detail & Related papers  (2025-09-23T14:42:32Z)
• CogDual: Enhancing Dual Cognition of LLMs via Reinforcement Learning with Implicit Rule-Based Rewards [53.36917093757101]
  Role-Playing Language Agents (RPLAs) have emerged as a significant application direction for Large Language Models (LLMs)<n>We introduce textbfCogDual, a novel RPLA adopting a textitcognize-then-respond reasoning paradigm.<n>By jointly modeling external situational awareness and internal self-awareness, CogDual generates responses with improved character consistency and contextual alignment.
  arXiv  Detail & Related papers  (2025-07-23T02:26:33Z)
• $φ^{\infty}$: Clause Purification, Embedding Realignment, and the Total Suppression of the Em Dash in Autoregressive Language Models [0.0]
  We identify a critical vulnerability in autoregressive transformer language models where the em dash token induces semantic drift.<n>We propose a novel solution combining symbolic clause purification via the phi-infinity operator with targeted embedding matrix.
  arXiv  Detail & Related papers  (2025-06-22T18:27:39Z)
• Membership Inference Attacks on Sequence Models [23.528760822574924]
  Sequence models, such as Large Language Models (LLMs) and autoregressive image generators, have a tendency to memorize and inadvertently leak sensitive information.<n>We argue that effectively measuring privacy leakage in sequence models requires leveraging the correlations inherent in sequential generation.
  arXiv  Detail & Related papers  (2025-06-05T15:13:57Z)
• AlignRAG: Leveraging Critique Learning for Evidence-Sensitive Retrieval-Augmented Reasoning [61.28113271728859]
  RAG has become a widely adopted paradigm for enabling knowledge-grounded large language models (LLMs)<n>Standard RAG pipelines often fail to ensure that model reasoning remains consistent with the evidence retrieved, leading to factual inconsistencies or unsupported conclusions.<n>In this work, we reinterpret RAG as Retrieval-Augmented Reasoning and identify a central but underexplored problem: textitReasoning Misalignment.
  arXiv  Detail & Related papers  (2025-04-21T04:56:47Z)
• Deliberative Alignment: Reasoning Enables Safer Language Models [64.60765108418062]
  We introduce Deliberative Alignment, a new paradigm that teaches the model safety specifications and trains it to explicitly recall and accurately reason over the specifications before answering.<n>We used this approach to align OpenAI's o-series models, and achieved highly precise adherence to OpenAI's safety policies, without requiring human-written chain-of-thoughts or answers.
  arXiv  Detail & Related papers  (2024-12-20T21:00:11Z)
• Fundamental Limitations of Alignment in Large Language Models [16.393916864600193]
  An important aspect in developing language models that interact with humans is aligning their behavior to be useful and unharmful. This is usually achieved by tuning the model in a way that enhances desired behaviors and inhibits undesired ones. We propose a theoretical approach called Behavior Expectation Bounds (BEB) which allows us to formally investigate several inherent characteristics and limitations of alignment in large language models.
  arXiv  Detail & Related papers  (2023-04-19T17:50:09Z)
• Policy Smoothing for Provably Robust Reinforcement Learning [109.90239627115336]
  We study the provable robustness of reinforcement learning against norm-bounded adversarial perturbations of the inputs. We generate certificates that guarantee that the total reward obtained by the smoothed policy will not fall below a certain threshold under a norm-bounded adversarial of perturbation the input.
  arXiv  Detail & Related papers  (2021-06-21T21:42:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.