Tuning-Free Accountable Intervention for LLM Deployment -- A Metacognitive Approach

CLEAR enables LLMs to self-identify and correct errors via concept-specific sparse subnetworks without tuning, improving deployment accountability.

Zhen Wah Tan, Jie Peng, Tianlong Chen, Huan Liu

Published Mar 8, 2024 · 3 citations

Don’t Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaboration

Multi-LLM collaboration detects knowledge gaps to make LLMs abstain from wrong answers instead of hallucinating.

Shangbin Feng, Weijia Shi, Yike Wang, Wenxuan Ding and 2 more

Published 2024 · 37 citations

Teaching LLMs to Abstain across Languages via Multilingual Feedback

Multilingual feedback teaches LLMs to abstain from answering in low-resource languages and improves cross-lingual abstention without degrading performance.

Shangbin Feng, Weijia Shi, Yike Wang, Wenxuan Ding and 5 more

Published 2024 · 4 citations

CHORD: Cross-Model Hallucination Detection via Relational Graph Discrimination

Yongxin Deng, Zhen Fang, Guansong Pang, Sharon Li and 1 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Beyond Uniform Detection: Adaptive Hallucination Detection for RAG Across Response Regimes

Jungwuk Park, Sejong Ryu, Jy-yong Sohn, Jaekyun Moon

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

When Should LLMs Be Less Specific? Selective Abstraction for Reliable Long-Form Text Generation

Shani Goren, Ido Galil, Ran El-Yaniv

Paris Poster Session 1, Wed, Dec 9, 12:30 PM–2:30 PM, Paris Poster Hall · Published 2026

When Copying Is Hard: Copy-Constrained Decoding for Exact Span Reproduction

Jinghui Zhang, Lang Gao, Zongfang Liu, Ruihong Zeng and 4 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Feeling of Knowing in Large Language Models

Zichuan Fu, Xian Wu, Jingtong Gao, Wenlin Zhang and 8 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

The Labeling Problem in Hallucination Detection Benchmarks: An Empirical Evaluation

Jorma Valjakka, Juhani Kivimäki, Juha Mylläri, Jukka K Nurminen

Paris Poster Session 5, Fri, Dec 11, 11:30 AM–1:30 PM, Paris Poster Hall · Published 2026

Why Are LLMs Confidently Wrong? Correcting Overconfident Errors via Causal Head Intervention

Jing Ren, Bowen Li, Ziqi Xu, Xuechao Yang and 1 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Rethinking in Spikes: Mitigating Hallucinations in MDLMs with Step-Aware Decoding

Zhongxing Xu, Zhonghua Wang, Zhe Qian, Shiyan Su and 8 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Large Language Model Failures from Hallucination to Homogenization Are Different Facets of Miscalibration

Tiancheng Hu, Caiqi Zhang, Dirk Hovy, Nigel Collier

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Shared Truth: Emergent Truth Properties in Large Language Models via Heterogeneous Injection-based Transfer

Isaiah Freeman, Joed Ngangmeni

Atlanta Poster Session 6, Fri, Dec 11, 4:30 PM–7:30 PM, Hall C1 · Published 2026

PLLS-CP: Unsupervised Hallucination Detection with Layer Selection and Cross-Domain Conformal Guarantees

Vladislav Tsvelenev, Maxim Ryndin

Paris Poster Session 1, Wed, Dec 9, 12:30 PM–2:30 PM, Paris Poster Hall · Published 2026

Alleviating Hallucination with Training-Free Uncertainty-Guided Steering

Litian Liu, Yubing Jian, Qiqi Hou, Reza Pourreza and 4 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

SOPO: Socratic Guided Policy Optimization for Span-Level Hallucination Detection

Jiawei Dong, Zilong Bai, Pengtian Zhu, Peng Zhou

Paris Poster Session 5, Fri, Dec 11, 11:30 AM–1:30 PM, Paris Poster Hall · Published 2026

Overthinking as a Symptom of Knowledge Conflict: Understanding and Detecting LLM's Hallucinations in Retrieval-Augmented Question Answering

Zhihua Wen, Li Hao, Zhiliang Tian, Zhizhao Liu and 3 more

Paris Poster Session 6, Fri, Dec 11, 2:30 PM–4:30 PM, Paris Poster Hall · Published 2026

Invisible Ink, Visible Lies: How Production Watermarking Causes LLMs to Hallucinate

Haocheng Ye, Aoting Hu, Xinwei Zhang, Xunzhu Tang and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Geometry-Calibrated Conformal Abstention for Language Models

Conformal Abstention uses geometry-calibrated confidence to decide abstention with finite-sample correctness guarantees, achieving 75% conditional correctness.

Rui Xu, Yi Chen, Sihong Xie, Hui Xiong

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation

BALTO applies balanced token-level credit assignment to mitigate LLM hallucinations by redistributing probability from unsupported to faithful content, outperforming response-level methods on faithfulness benchmarks.

Ning Li, Zixuan Guo, Yan Xu, Wenbo Fei and 6 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Entropy Distribution as a Fingerprint for Hallucinations in Generative Models

Token-level entropy distributions fingerprint hallucinations, and the single-pass Calibrated Entropy Score achieves multi-pass detection accuracy with formal guarantees.

Mattia Jacopo Villani, Pranav Deshpande, Akshay Seshadri, Romina Yalovetzky and 1 more

Atlanta Poster Session 4, Thu, Dec 10, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Scalable Token-Level Hallucination Detection in Large Language Models

TokenHD trains token-level hallucination detectors via scalable synthetic data and importance weighting, with small models outperforming larger reasoning models and scaling consistently.

Rui Min, Tianyu Pang, Chao Du, Minhao Cheng and 1 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

KG-Guard: Graph-Based Hallucination Detection for Knowledge Base Question Answering

KG-Guard detects KBQA hallucinations via graph-based answer-node classification, achieving 82, 87 F1 with 305x fewer parameters and improving downstream accuracy 13, 15 points via iterative refinement.

Albert Sawczyn, Piotr Bielak, Tomasz Kajdanowicz

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Negation Neglect: When models fail to learn negations in training

Fine-tuning LLMs on documents that flag claims as false makes them believe those claims, with belief rates jumping from 2.5% to 88.6%, though local negation phrasing largely prevents it.

Harry Mayne, Lev McKinney, Jan Dubiński, Adam Karvonen and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Kernel Token Contradiction: a Fast and Principled Approach for LLM Claim Uncertainty Quantification

Kernel Token Contradiction uses a token contradiction kernel with von Neumann entropy for fast, accurate LLM claim-level uncertainty quantification. It achieves over 8.2x speedups versus GPU cross-encoders and 65x versus CPU baselines while matching or exceeding accuracy, especially in high-precisio

Jérémie Dentan, Alexi Canesse, Mahammed El Sharkawy, Sonia Vanier

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Learning When to Trust LLM Priors: A Validated Framework for Semantic Prior Integration

Statsformer validates LLM semantic priors via out-of-fold calibration to adaptively integrate them across diverse predictors, guaranteeing performance at least as good as the best convex combination of candidates.

Erica Zhang, Naomi Sagan, Danny Tse, Fangzhao Zhang and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

The Future of Facts: Tracing the Factual Generation-Verification Gap

Verification of facts is consistently learned before generation, resists continual learning better, and leaves models verifying both old and new answers after updates.

Tim R Davidson, Anja Surina, Caglar Gulcehre

Paris Poster Session 4, Thu, Dec 10, 5:30 PM–7:30 PM, Paris Poster Hall · Published 2026

Tracing the Cascade: A Topology-Aware Evaluation Framework for Scientific Agent Hallucinations

SCHEMA evaluates scientific agent hallucinations via topology-aware diagnostics, showing errors cluster at connected knowledge hubs and correct answers often stem from flawed reasoning.

Xinshun Feng, Ziqi Miao, Lijun Li, Jing Shao

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Uncertainty Quantification for Large Language Diffusion Models

Lightweight zero-shot uncertainty signals from LLDM denoising dynamics achieve sampling-level hallucination detection at up to 100x lower cost.

Artem Vazhentsev, Vladislav Smirnov, David Li, Maxim Panov and 2 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Hallucination as Commitment Failure: Larger LLMs Misfire Despite Knowing the Answer

Larger instruction-tuned LLMs increasingly hallucinate despite knowing correct answers because sharpening commitment disperses probability across surface forms rather than concentrating it.

Jewon Yeom, Jaewon Sok, Heejun Kim, Seonghyeon Park and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026