REVIEW 7 cited by
Trust Me, I'm Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answer
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Prior work on large language model (LLM) hallucinations has associated them with model uncertainty or inaccurate knowledge. In this work, we define and investigate a distinct type of hallucination, where a model can consistently answer a question correctly, but a seemingly trivial perturbation, which can happen in real-world settings, causes it to produce a hallucinated response with high certainty. This phenomenon, which we dub CHOKE (Certain Hallucinations Overriding Known Evidence), is particularly concerning in high-stakes domains such as medicine or law, where model certainty is often used as a proxy for reliability. We show that CHOKE examples are consistent across prompts, occur in different models and datasets, and are fundamentally distinct from other hallucinations. This difference leads existing mitigation methods to perform worse on CHOKE examples than on general hallucinations. Finally, we introduce a probing-based mitigation that outperforms existing methods on CHOKE hallucinations. These findings reveal an overlooked aspect of hallucinations, emphasizing the need to understand their origins and improve mitigation strategies to enhance LLM safety. The code is available at https://github.com/technion-cs-nlp/Trust_me_Im_wrong .
Forward citations
Cited by 7 Pith papers
-
A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations
A single linear direction in an observer model's residual stream detects contextual hallucinations, transfers across models and datasets, and causally steers generation hallucination rates.
-
$\Sigma$-Mem: An Online Reliability Memory for LLM-based Multi-Agent Systems
Online symmetric reliability memory for LLM multi-agent systems accumulates bounded competence and peer-relationship evidence and supports steering, routing, and weighted voting without retraining.
-
Confidently Wrong: Detecting Hallucinations in Financial Question Answering from LLM Internal States
Among 8/8 self-consistent answers on FinQA, residual-stream probes detect wrong answers at 0.68–0.77 AUROC versus 0.55–0.63 for the best cheap output baselines across three 8–9B models.
-
Reliability Scales Inversely: Hallucinations Snowball Faster in Bigger Language Models
Hallucination in LLMs is driven by an oracle-invisible “decoding risk” term that grows with scale and causally compounds errors within a response.
-
Benchmarking LLM-Assisted Blue Teaming via Standardized Threat Hunting
Standardized modular threat-hunting workflows (CyberTeam) improve LLM performance on blue team tasks compared to open-ended ICL, CoT, and ToT prompting across 30 tasks and 452k samples.
-
The Future is Agentic: Definitions, Perspectives, and Open Challenges of Multi-Agent Recommender Systems
A framework for agentic recommender systems plus a pilot study showing multi-agent pipelines beat a single-shot LLM only on high-diversity user histories.
-
Tractable Asymmetric Verification for Large Language Models via Deterministic Replicability
An LLM output can be verified by regenerating a few randomly chosen segments under identical hardware, with a tunable detection probability and 12.4x speedup over full regeneration.
Discussion (0). Sign in to comment.