Models delayed verification in multi-agent LLMs as graph consensus, derives stability thresholds (inverse golden ratio for delay two) via grounded Laplacian, and gives a supermodular greedy rule for corrector placement; experiments on five models confirm dose-delay oscillations.
hub
arXiv preprint arXiv:2305.13534 , year=
19 Pith papers cite this work, alongside 71 external citations. Polarity classification is still indexing.
19
Pith papers citing it
71
external citations · Pith
abstract
A major risk of using language models in practical applications is their tendency to hallucinate incorrect statements. Hallucinations are often attributed to knowledge gaps in LMs, but we hypothesize that in some cases, when justifying previously generated hallucinations, LMs output false claims that they can separately recognize as incorrect. We construct three question-answering datasets where ChatGPT and GPT-4 often state an incorrect answer and offer an explanation with at least one incorrect claim. Crucially, we find that ChatGPT and GPT-4 can identify 67% and 87% of their own mistakes, respectively. We refer to this phenomenon as hallucination snowballing: an LM over-commits to early mistakes, leading to more mistakes that it otherwise would not make.
hub tools
representative citing papers
LinAlg-Bench shows LLMs switch from execution errors to computational abandonment and structured fabrication at 4x4 matrix scale, indicating a working memory limit rather than knowledge gaps.
StrucTab achieves SOTA table parsing performance by unifying structural subtasks through sequential reasoning and using decomposed RL rewards in Uni-TabRL, plus a new TableVerse-5K benchmark.
PseudoBench shows current LLM agents produce persuasive pseudoscientific reports with near-zero refusal rates and at most 27.4% resistance.
Agent systems lose uncertainty at decision handoffs, causing downstream over-trust; the paper proposes latent uncertainty as a carrier to preserve pre-commitment fragility across interfaces.
Agent-directed tree search improves LLM performance on Lean formal verification tasks, with context-based orchestration solving more intermediate specs at lower token cost than baseline agents.
APT-Agent automates penetration testing with LLMs using rectification and memory modules, achieving 84.29% end-to-end success on Metasploitable 2 versus lower rates for baselines.
OEP poisons self-evolving LLM agents by constructing clean edge-case experiences that appear locally valid yet cause harmful over-generalization during reflection, achieving over 50% attack success rate on GPT-4o agents across three domains.
RAG-Pref is a training-free RAG-based alignment technique that conditions LLMs on contrastive preference samples during inference, yielding over 3.7x average improvement in agentic attack refusals when combined with offline methods across five LLMs.
Reducing precision from 16-bit to 8/4-bit in multi-hop reasoning creates a quantization trap that raises net energy consumption and degrades accuracy, breaking linear scaling laws.
The method aggregates multiple hallucination evaluation scores via conformal p-values to enable calibrated detection with controlled false alarm rates across LLMs and datasets.
LURE reduces object hallucination in LVLMs by 23% via post-hoc revision informed by co-occurrence, uncertainty, and text position analysis.
Chain-of-Verification reduces hallucinations in large language models by drafting responses, planning independent verification questions, answering them separately, and generating a final verified output.
A game-framed multi-agent system synthesizes large chemistry CoT/QA corpora and trains OmniChem-7B to near GPT-4o-mini performance with a large reported drop in hallucinations.
Enforcing structured reflection via Outlines-based constrained decoding on an 8B LLM triggers structure snowballing instead of better self-correction, producing near-perfect syntax but persistent semantic errors and revealing an alignment tax.
A survey that compiles and taxonomizes more than 32 existing hallucination mitigation techniques for LLMs while analyzing their challenges and limitations.
A survey of emerging AI agent architectures that organizes single and multi-agent designs around reasoning, planning, tool use, communication, and reflection phases.
A survey classifying hallucination phenomena specific to large foundation models, establishing evaluation criteria, examining mitigation strategies, and discussing future directions.
The paper identifies social and ethical risks from unguarded use of general-purpose LLMs in Canadian newcomer settlement and advocates for AI literacy programs plus customized models with human oversight.
citing papers explorer
-
Delayed Verification Destabilizes Multi-Agent LLM Belief: Instability Thresholds and Optimal Corrector Placement
Models delayed verification in multi-agent LLMs as graph consensus, derives stability thresholds (inverse golden ratio for delay two) via grounded Laplacian, and gives a supermodular greedy rule for corrector placement; experiments on five models confirm dose-delay oscillations.
-
LinAlg-Bench: A Forensic Benchmark Revealing Structural Failure Modes in LLM Mathematical Reasoning
LinAlg-Bench shows LLMs switch from execution errors to computational abandonment and structured fabrication at 4x4 matrix scale, indicating a working memory limit rather than knowledge gaps.
-
StrucTab: A Structured Optimization Framework for Table Parsing
StrucTab achieves SOTA table parsing performance by unifying structural subtasks through sequential reasoning and using decomposed RL rewards in Uni-TabRL, plus a new TableVerse-5K benchmark.
-
PseudoBench: Measuring How Agentic Auto-Research Fuels Pseudoscience
PseudoBench shows current LLM agents produce persuasive pseudoscientific reports with near-zero refusal rates and at most 27.4% resistance.
-
Confidence Laundering in Agent Systems: Why Uncertainty Needs a Latent Carrier
Agent systems lose uncertainty at decision handoffs, causing downstream over-trust; the paper proposes latent uncertainty as a carrier to preserve pre-commitment fragility across interfaces.
-
Automating Formal Verification with Agent-Guided Tree Search
Agent-directed tree search improves LLM performance on Lean formal verification tasks, with context-based orchestration solving more intermediate specs at lower token cost than baseline agents.
-
APT-Agent: Automated Penetration Testing using Large Language Models
APT-Agent automates penetration testing with LLMs using rectification and memory modules, achieving 84.29% end-to-end success on Metasploitable 2 versus lower rates for baselines.
-
OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences
OEP poisons self-evolving LLM agents by constructing clean edge-case experiences that appear locally valid yet cause harmful over-generalization during reflection, achieving over 50% attack success rate on GPT-4o agents across three domains.
-
Leveraging RAG for Training-Free Alignment of LLMs
RAG-Pref is a training-free RAG-based alignment technique that conditions LLMs on contrastive preference samples during inference, yielding over 3.7x average improvement in agentic attack refusals when combined with offline methods across five LLMs.
-
The Quantization Trap: Breaking Linear Scaling Laws in Multi-Hop Reasoning
Reducing precision from 16-bit to 8/4-bit in multi-hop reasoning creates a quantization trap that raises net energy consumption and degrades accuracy, breaking linear scaling laws.
-
Principled Detection of Hallucinations in Large Language Models via Multiple Testing
The method aggregates multiple hallucination evaluation scores via conformal p-values to enable calibrated detection with controlled false alarm rates across LLMs and datasets.
-
Analyzing and Mitigating Object Hallucination in Large Vision-Language Models
LURE reduces object hallucination in LVLMs by 23% via post-hoc revision informed by co-occurrence, uncertainty, and text position analysis.
-
Chain-of-Verification Reduces Hallucination in Large Language Models
Chain-of-Verification reduces hallucinations in large language models by drafting responses, planning independent verification questions, answering them separately, and generating a final verified output.
-
Game Theory Driven Multi-Agent Framework Mitigates Language Model Hallucination
A game-framed multi-agent system synthesizes large chemistry CoT/QA corpora and trains OmniChem-7B to near GPT-4o-mini performance with a large reported drop in hallucinations.
-
From Hallucination to Structure Snowballing: The Alignment Tax of Constrained Decoding in LLM Reflection
Enforcing structured reflection via Outlines-based constrained decoding on an 8B LLM triggers structure snowballing instead of better self-correction, producing near-perfect syntax but persistent semantic errors and revealing an alignment tax.
-
A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Models
A survey that compiles and taxonomizes more than 32 existing hallucination mitigation techniques for LLMs while analyzing their challenges and limitations.
-
The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey
A survey of emerging AI agent architectures that organizes single and multi-agent designs around reasoning, planning, tool use, communication, and reflection phases.
-
A Survey of Hallucination in Large Foundation Models
A survey classifying hallucination phenomena specific to large foundation models, establishing evaluation criteria, examining mitigation strategies, and discussing future directions.
-
Social and Ethical Risks Posed by General-Purpose LLMs for Settling Newcomers in Canada
The paper identifies social and ethical risks from unguarded use of general-purpose LLMs in Canadian newcomer settlement and advocates for AI literacy programs plus customized models with human oversight.