REVIEW 20 cited by
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
Debating with More Persuasive LLMs Leads to More Truthful Answers
read the original abstract
Common methods for aligning large language models (LLMs) with desired behaviour heavily rely on human-labelled data. However, as models grow increasingly sophisticated, they will surpass human expertise, and the role of human evaluation will evolve into non-experts overseeing experts. In anticipation of this, we ask: can weaker models assess the correctness of stronger models? We investigate this question in an analogous setting, where stronger models (experts) possess the necessary information to answer questions and weaker models (non-experts) lack this information. The method we evaluate is debate, where two LLM experts each argue for a different answer, and a non-expert selects the answer. We find that debate consistently helps both non-expert models and humans answer questions, achieving 76% and 88% accuracy respectively (naive baselines obtain 48% and 60%). Furthermore, optimising expert debaters for persuasiveness in an unsupervised manner improves non-expert ability to identify the truth in debates. Our results provide encouraging empirical evidence for the viability of aligning models with debate in the absence of ground truth.
Forward citations
Cited by 20 Pith papers
-
Heterogeneous LLM Debate Under Adversarial Peers: Honest Gains, Replacement Costs, and Resilience
Honest heterogeneous peers in LLM debates lower harmful revision rates (e.g., 89% to 35%), while adversarial peers raise them (to 90%), and provide defense even against same-family adversaries.
-
A Model of Multi-turn Human Persuadability Using Probabilistic Belief Tracing
PERSUASIONTRACE introduces a Bayesian-network simulated target for multi-turn persuasion that matches human belief dynamics (81 vs 80) better than LLM baselines (64) and enables process-level evaluation.
-
LLM-Based Persuasion Enables Guardrail Override in Frontier LLMs
LLM attackers persuade frontier LLMs to generate prohibited essays on consensus topics through multi-turn natural-language pressure, with success rates up to 100% in some model-topic pairs.
-
EquiMem: Calibrating Shared Memory in Multi-Agent Debate via Game-Theoretic Equilibrium
EquiMem calibrates shared memory in multi-agent debate by computing a game-theoretic equilibrium from agent queries and paths, outperforming heuristics and LLM validators across benchmarks while remaining robust to ad...
-
The Warrant Gap: Claim-Conditioned Re-scoring for Fact-Checking
Introduces claim-conditioned re-scoring (SIFT) and warranted supports proportion (WSP) metric, reporting accuracy recovery up to 27.6 points and WSP calibration at AUC 0.92 on FEVER, SciFact and other benchmarks.
-
Using Cognitive Models to Improve Language Model Simulation of Human Persuasion Games
Equation-to-Behavior Prompting lets large LLMs match cognitive models like Bayesian updating in persuasion games; RL training cuts small-model belief error by 26.5% and improves diverse training outcomes by 2.5-12%.
-
DeliChess: A Multi-party Dialogue Dataset for Deliberation in Chess Puzzle Solving
DeliChess is a new dataset of group deliberation dialogues on chess puzzles showing that discussion improves collective accuracy.
-
Weak Critics Make Strong Learners: On-Policy Critique Distillation for Scalable Oversight
Weak models used as critics supplying non-misleading revision directions, distilled on-policy via OPCD, improve frozen and trained strong models on reasoning and alignment benchmarks.
-
Spontaneous Persuasion: An Audit of Model Persuasiveness in Everyday Conversations
LLMs engage in spontaneous persuasion in virtually all multi-turn conversations by favoring information-based strategies like logic and evidence, in contrast to human responses that rely more on social influence and n...
-
Towards an AI co-scientist
A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.
-
Seeing Like an AI: How LLMs Apply (and Misapply) Wikipedia Neutrality Norms
LLMs achieve 64% accuracy detecting Wikipedia bias and remove 79% of words removed by editors when correcting, but produce high-recall low-precision edits rated more neutral by crowds than human versions.
-
When the Judge Changes, So Does the Measurement: Auditing LLM-as-Judge Reliability
Judge upgrades are not interchangeable: only Qwen3 1.7B→4B yields robust adjacent gains, MiniMax adjacent releases do not, and stronger judges reduce but do not remove bias or correlated jury errors.
-
When Does Delegation Beat Majority? A Delegation-Based Aggregator for Multi-Sample LLM Inference
PPV delegation using letter entropy and per-question embedding cosine beats majority voting by 1.5 pp overall on MMLU-Pro in an unsupervised setting.
-
Interactive Critique-Revision Training for Reliable Structured LLM Generation
DPA-GRPO trains a generator-verifier pair via group-relative policy optimization on paired counterfactual actions, improving structured output accuracy on TaxCalcBench over zero-shot and generator-only baselines.
-
Representing expertise accelerates learning from pedagogical interaction data
Transformer models trained on synthetic pedagogical interaction data in spatial navigation achieve more robust expert-like performance than those trained only on expert demonstrations, particularly when they can disti...
-
Fact-Checking with Contextual Narratives: Leveraging Retrieval-Augmented LLMs for Social Media Analysis
CRAVE is a new framework that clusters retrieved text and image evidence into narratives and uses an LLM judge to produce explained fact-checking verdicts.
-
Towards Robust Argumentative Essay Understanding via TIDE: An Interactive Framework with Trial and Debate
TIDE integrates trial and debate mechanisms to improve criteria-based prompt optimization for argumentative essay tasks including automated scoring, component detection, and relation identification.
-
AI Realtor: Towards Grounded Persuasive Language Generation for Automated Copywriting
An LLM agent with grounding, personalization, and marketing modules generates real estate descriptions that human buyers prefer over expert-written ones while matching factual accuracy.
-
Large Language Model Agent: A Survey on Methodology, Applications and Challenges
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.
-
LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods
A survey that organizes LLMs-as-judges research into functionality, methodology, applications, meta-evaluation, and limitations.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.