NeuroGenesysThe research we build on.
1,514 papers by other researchers that our work cites, from continual learning and verification to language models, agents and hardware. Each entry links to the paper on arXiv and to where we use it. 301 are cited on published pages of this site; the rest in the research notes behind our studies. Our own papers are in the archive.
- 2026-09arXiv 2609.17686The Missing "I Don't Know": Why Three Reasoning-Reliability Findings Converge on Calibrated Abstention
Srijith Ravikumar · Machine learning
Three recent results describe what look like unrelated LLM reliability problems. Yin et al. (2026) show reasoning RL collapses tool-reliability representations. Suleymanov et al.
- 2026-09arXiv 2609.07395Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Study
Moule Lin, Qizhen Lan, Shuhao Guan et al. · Retrieval & data
Large language models (LLMs) are no longer deployed only for single-turn conversation but increasingly act as agents that plan, call tools, retrieve evidence, maintain memory, and interact over long horizons, often together with other agents through multi-turn conversations.
- 2026-09arXiv 2609.04445Conformity Breaks Conformal Prediction
Yibo Hu, Hanyu Su · Machine learning
A conformal certificate can be valid when an LLM answers alone and invalid when the same LLM sees peers that unanimously assert a wrong answer. The question is unchanged; the model's score for the correct answer changes.
- 2026-07arXiv 2608.00301Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning
Xujun Che, Yuchen Yuan, Weida Zhao et al. · Machine learning
Error-penalized scoring rules ( for a correct answer, for a wrong one, $0$ for abstaining) are increasingly prescribed against hallucination: a rational agent facing such a rule answers exactly when its correctness probability exceeds Chow's threshold .
- 2026-07arXiv 2607.16799Diagnosing Correctness Probes under Self-Judgement Confounding
Yi-Long Lu · Language models & NLP
Hidden-state readouts can predict whether language-model outputs are correct, but objective correctness (OC) usually agrees with the model's own self-judgement (SJ), leaving the decoded signal semantically ambiguous.
- 2026-07arXiv 2607.13568Graded Entity-Familiarity Readouts in Language Models: Polish Adaptation, Cross-Language Robustness, and Refusal Steering
Grzegorz Brzezinka · Language models & NLP
Can a language model estimate its familiarity with an entity before generating an answer?
- 2026-07arXiv 2607.08065When LLMs Agree, Are They Right? Auditing Self-Consistency and Cross-Model Agreement as Confidence Signals
Kaihua Ding · AI & agents
, 2017) panels of judges. These systems share a key assumption: that consistency -- agreement among judges, or among a model's own samples -- indicates correctness.
- 2026-05arXiv 2605.26242Can LLMs Introspect? A Reality Check
Shashwat Singh, Tal Linzen, Shauli Ravfogel · AI & agents
Can large language models detect and report their own internal states? A number of recent studies have argued that they can. Drawing on lessons from human metacognition research, we argue that this conclusion may be premature.
- 2026-04arXiv 2606.11211Calibration Drift Under Reasoning: How Chain-of-Thought Budgets Induce Overconfidence in Large Language Models
Prakul Sunil Hiremath, Harshit R. Hiremath · Language models & NLP
The ability of large language models (LLMs) to express calibrated uncertainty is important for safe deployment. Chain-of-thought (CoT) reasoning is widely used to improve accuracy and reliability, but its effect on calibration is not fully understood.
- 2026-04arXiv 2604.15109IUQ: Interrogative Uncertainty Quantification for Long-Form Large Language Model Generation
Haozhi Fan, Jinhao Duan, Kaidi Xu · Language models & NLP
Despite the rapid advancement of Large Language Models (LLMs), uncertainty quantification in LLM generation is a persistent challenge.
- 2026-04arXiv 2604.05358LatentAudit: Real-Time White-Box Faithfulness Monitoring for Retrieval-Augmented Generation with Verifiable Deployment
Zhe Yu, Wenpeng Xing, Meng Han · AI & agents
Retrieval-augmented generation (RAG) mitigates hallucination but does not eliminate it: a deployed system must still decide, at inference time, whether its answer is actually supported by the retrieved evidence.
- 2026-03arXiv 2603.21396Mechanisms of Introspective Awareness
Uzay Macar, Li Yang, Atticus Wang et al. · Machine learning
" We investigate the mechanisms underlying this capability in open-weights models. First, we find that it is behaviorally robust: models detect injected steering vectors at moderate rates with 0% false positives across diverse prompts and dialogue formats.
- 2026-02arXiv 2602.20031Latent Introspection: Models Can Detect Prior Concept Injections
Theia Pearson-Vogel, Martin Vanek, Raymond Douglas et al. · AI & agents
We uncover a latent capacity for introspection in a Qwen 32B model, demonstrating that the model can detect when concepts have been injected into its earlier context and identify which concept was injected.
- 2026-02arXiv 2602.13093Consistency of Large Reasoning Models Under Multi-Turn Attacks
Yubo Li, Ramayya Krishnan, Rema Padman · AI & agents
Large reasoning models with reasoning capabilities achieve state-of-the-art performance on complex tasks, but their robustness under multi-turn adversarial pressure remains underexplored.
- 2026-01arXiv 2601.19921Demystifying Multi-Agent Debate: The Role of Confidence and Diversity
Xiaochen Zhu, Caiqi Zhang, Yizhou Chi et al. · Language models & NLP
Multi-agent debate (MAD) is widely used to improve large language model (LLM) performance through test-time scaling, yet recent work shows that vanilla MAD often underperforms simple majority vote despite higher computational cost.
- 2025-12arXiv 2512.19920Mitigating LLM Hallucination via Behaviorally Calibrated Reinforcement Learning
Jiayun Wu, Jiashuo Liu, Zhiyuan Zeng et al. · Machine learning
LLM deployment in critical domains is currently impeded by persistent hallucinations--generating plausible but factually incorrect assertions.
- 2025-12arXiv 2512.01725Beware of Reasoning Overconfidence: Pitfalls in the Reasoning Process for Multi-solution Tasks
Jiannan Guan, Qiguang Chen, Libo Qin et al. · Language models & NLP
Large Language Models (LLMs) excel in reasoning tasks requiring a single correct answer, but they perform poorly in multi-solution tasks that require generating comprehensive and diverse answers.
- 2025-10arXiv 2510.12040Uncertainty Quantification for Hallucination Detection in Large Language Models: Foundations, Methodology, and Future Directions
Sungmin Kang, Yavuz Faruk Bakman, Duygu Nur Yaldiz et al. · Language models & NLP
The rapid advancement of large language models (LLMs) has transformed the landscape of natural language processing, enabling breakthroughs across a wide range of areas including question answering, machine translation, and text summarization.
- 2025-10arXiv 2510.15977Bolster Hallucination Detection via Prompt-Guided Data Augmentation
Wenyun Li, Zheng Zhang, Dongmei Jiang et al. · Machine learning
Large language models (LLMs) have garnered significant interest in AI community. Despite their impressive generation capabilities, they have been found to produce misleading or fabricated information, a phenomenon known as hallucinations.
- 2025-10arXiv 2510.05566Domain-Shift-Aware Conformal Prediction for Large Language Models
Zhexiao Lin, Yuanyuan Li, Neeraj Sarna et al. · Machine learning
Large language models have achieved impressive performance across diverse tasks. However, their tendency to produce overconfident and factually incorrect outputs, known as hallucinations, poses risks in real-world applications.
- 2025-09arXiv 2509.25760TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
Zhepei Wei, Xiao Yang, Kai Sun et al. · Language models & NLP
While large language models (LLMs) have demonstrated strong performance on factoid question answering, they are still prone to hallucination and untruthful responses, particularly when tasks demand information outside their parametric knowledge.
- 2025-09arXiv 2509.15631Sparse-Autoencoder-Guided Internal Representation Unlearning for Large Language Models
Tomoya Yamashita, Akira Ito, Yuuki Yamanaka et al. · Language models & NLP
As large language models (LLMs) are increasingly deployed across various applications, privacy and copyright concerns have heightened the need for more effective LLM unlearning techniques.
- 2025-09arXiv 2509.19372Representation-based Broad Hallucination Detectors Fail to Generalize Out of Distribution
Zuzanna Dubanowska, Maciej Żelaszczyk, Michał Brzozowski et al. · Machine learning
We critically assess the efficacy of the current SOTA in hallucination detection and find that its performance on the RAGTruth dataset is largely driven by a spurious correlation with data.
- 2025-09arXiv 2509.04664Why Language Models Hallucinate
Adam Tauman Kalai, Ofir Nachum, Santosh S. Vempala et al. · Language models & NLP
Like students facing hard exam questions, large language models sometimes guess when uncertain, producing plausible yet incorrect statements instead of admitting uncertainty.
- 2025-08arXiv 2509.03531Real-Time Detection of Hallucinated Entities in Long-Form Generation
Oscar Obeso, Andy Arditi, Javier Ferrando et al. · Language models & NLP
Large language models are now routinely used in high-stakes applications where hallucinations can cause serious harm, such as medical consultations or legal advice.
- 2025-08arXiv 2508.17536Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?
Hyeong Kyu Choi, Xiaojin Zhu, Sharon Li · Language models & NLP
Multi-Agent Debate~(MAD) has emerged as a promising paradigm for improving the performance of large language models through collaborative reasoning. Despite recent advances, the key factors driving MAD's effectiveness remain unclear.
Cited inCoherence is selected, anchored, then pruned: a literature synthesis on logical coherence in LLM world modelsStable wrong beliefs evade every self-check: a literature synthesis on model self-knowledge and confident errorsStudy: Logical coherence of LLM world modelsStudy: Model self knowledge and confident errorsStudy: Orchestrated specialists and bridge synthesis - 2025-08arXiv 2508.15050Don't Think Twice! Over-Reasoning Impairs Confidence Calibration
Romain Lacombe, Kerrie Wu, Eddie Dilworth · AI & agents
Large Language Models deployed as question answering tools require robust calibration to avoid overconfidence., 2023) and expanding it to human and planetary health.
- 2025-07arXiv 2507.16488ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMs
Zhenliang Zhang, Xinyu Hu, Huixuan Zhang et al. · Language models & NLP
Large language models (LLMs) excel at various natural language processing tasks, but their tendency to generate hallucinations undermines their reliability.
- 2025-07arXiv 2507.16806Beyond Binary Rewards: Training LMs to Reason About Their Uncertainty
Mehul Damani, Isha Puri, Stewart Slocum et al. · Machine learning
When language models (LMs) are trained via reinforcement learning (RL) to generate natural language "reasoning chains", their performance improves on a variety of difficult question answering tasks.
- 2025-06arXiv 2506.18183Reasoning about Uncertainty: Do Reasoning Models Know When They Don't Know?
Zhiting Mei, Christina Zhang, Tenny Yin et al. · AI & agents
Reasoning language models have set state-of-the-art (SOTA) records on many challenging benchmarks, enabled by multi-step reasoning induced using reinforcement learning.
- 2025-06arXiv 2506.09038AbstentionBench: Reasoning LLMs Fail on Unanswerable Questions
Polina Kirichenko, Mark Ibrahim, Kamalika Chaudhuri et al. · AI & agents
For Large Language Models (LLMs) to be reliably deployed in both everyday and high-stakes domains, knowing when not to answer is equally critical as answering correctly.
- 2025-06arXiv 2506.07962Correlated Errors in Large Language Models
Elliot Kim, Avi Garg, Kenny Peng et al. · Language models & NLP
Diversity in training data, architecture, and providers is assumed to mitigate homogeneity in LLMs. However, we lack empirical evidence on whether different LLMs differ meaningfully.
- 2025-05arXiv 2505.16170When Do LLMs Admit Their Mistakes? Understanding The Role Of Model Belief In Retraction
Yuqing Yang, Robin Jia · Language models & NLP
, spontaneously and immediately acknowledge errors in their previously generated false assertions. Using model-specific testbeds, we find that while LLMs are capable of retraction, they do so only rarely, even when they can recognize their mistakes when asked in a separate interaction.
Cited inCoherence is selected, anchored, then pruned: a literature synthesis on logical coherence in LLM world modelsStable wrong beliefs evade every self-check: a literature synthesis on model self-knowledge and confident errorsStudy: Logical coherence of LLM world modelsStudy: Model self knowledge and confident errors - 2025-05arXiv 2505.17126Conformal Language Model Reasoning with Coherent Factuality
Maxon Rubin-Toles, Maya Gambhir, Keshav Ramji et al. · Language models & NLP
Language models are increasingly being used in important decision pipelines, so ensuring the correctness of their outputs is crucial.
Cited inCoherence is selected, anchored, then pruned: a literature synthesis on logical coherence in LLM world modelsStable wrong beliefs evade every self-check: a literature synthesis on model self-knowledge and confident errorsStudy: Logical coherence of LLM world modelsStudy: Model self knowledge and confident errors - 2025-05arXiv 2505.13988The Hallucination Tax of Reinforcement Finetuning
Linxin Song, Taiwei Shi, Jieyu Zhao · Language models & NLP
Reinforcement finetuning (RFT) has become a standard approach for enhancing the reasoning capabilities of large language models (LLMs). However, its impact on model trustworthiness remains underexplored.
- 2025-05arXiv 2505.14489Reasoning Models Better Express Their Confidence
Dongkeun Yoon, Seungone Kim, Sohee Yang et al. · AI & agents
Despite their strengths, large language models (LLMs) often fail to communicate their confidence accurately, making it difficult to assess when they might be wrong and limiting their reliability.
- 2025-04arXiv 2504.17550HalluLens: LLM Hallucination Benchmark
Yejin Bang, Ziwei Ji, Alan Schelten et al. · Language models & NLP
" These hallucinations undermine user trust and hinder the adoption of generative AI systems. Addressing hallucinations is essential for the advancement of LLMs.
- 2025-04arXiv 2504.06564Thinking Out Loud: Do Reasoning Models Know When They're Right?
Qingcheng Zeng, Weihao Xuan, Leyang Cui et al. · Language models & NLP
Large reasoning models (LRMs) have recently demonstrated impressive capabilities in complex reasoning tasks by leveraging increased test-time computation and exhibiting behaviors reminiscent of human-like self-reflection.
- 2025-04arXiv 2504.03579Hallucination Detection on a Budget: Efficient Bayesian Estimation of Semantic Entropy
Kamil Ciosek, Nicolò Felicioni, Sina Ghiassian · Machine learning
Detecting whether an LLM hallucinates is an important research challenge., 2024) of the distribution of generated sequences. We propose a new algorithm for doing that, with two main advantages.
- 2025-03arXiv 2503.15299Inside-Out: Hidden Factual Knowledge in LLMs
Zorik Gekhman, Eyal Ben David, Hadas Orgad et al. · Language models & NLP
This work presents a framework for assessing whether large language models (LLMs) encode more factual knowledge in their parameters than what they express in their outputs.
Built from every arXiv paper cited in our papers, posts, topic hubs, reports and research notes, with titles, authors, dates and abstracts from the arXiv API (updated 2026-11-12). Summaries are the opening sentences of each paper's own abstract, not our assessment; a citation means we used or checked the work, not that we endorse it. Fields follow each paper's primary arXiv category. Papers cited only outside arXiv (journals, proceedings) are not listed yet. Entries are generated from our citations, so a mistyped ID in a note would list the wrong paper; we remove those as we find them. Report a wrong entry.
