arXiv AI

Decomposing Wrong-Consensus Agreement in LLM Self-Consistency: A GPT-4.1 Case Study

The paper introduces a pluralistic agreement index, Gamma, to quantify how often wrong runs of large language models (LLMs) agree with the majority consensus. By decomposing Gamma into a mechanical component and a preference‑unexplained residual, the authors show that on GPT‑4.1 the mechanical part explains most of the agreement on multiple‑choice benchmarks but only about half on open‑domain tasks, revealing a residual bias that can cause self‑consistency to backfire on hard questions. The study provides a quantitative framework for understanding when majority voting over LLM samples improves or harms accuracy, without proposing new voting methods.

arXiv AI

Sep 1

Decomposing Wrong-Consensus Agreement in LLM Self-Consistency

The paper investigates the nature of agreement among repeated samples of large language models (LLMs), showing that strong agreement can arise even for incorrect answers. It introduces a pluralistic agreement index, Gamma, which is decomposed into a mechanical component driven solely by per‑case answer preferences and a residual component that captures preference‑unexplained agreement. Experiments on GPT‑4.1 and several open‑weight models demonstrate that mechanical agreement dominates in many settings, while the residual varies with benchmark type and sampling protocol.

By Lizhuo Zhang, Mengmeng Tang, Chenfeng Long, Xiaoyong Tang, Xiang Luo

llms benchmarks safety

arXiv AI

Aug 13

When Self-Consistency Backfires: Majority Vote Hurts the Majority of Hard Science Problems for Small LLMs

arXiv:2608. 11403v1 Announce Type: new Abstract: Self-consistency (SC) via majority vote is a widely used way to spend inference-time compute: sample N chains of thought, return the plurality answer.

By Utkarsh Bahuguna

llms benchmarks

arXiv AI

Jun 15

The Coin Flip Judge? Reliability and Bias in LLM-as-a-Judge Evaluation

arXiv:2606. 13685v1 Announce Type: cross Abstract: LLM-as-a-Judge is now widely used to rank model outputs, train reward models, and populate public leaderboards, but its run-to-run reliability remains under-characterized.

By Abel Yagubyan

llms safety

arXiv AI

Jul 10

When LLMs Agree, Are They Right? Auditing Self-Consistency and Cross-Model Agreement as Confidence Signals

arXiv:2607. 08065v1 Announce Type: new Abstract: LLM-as-judge (Zheng et al.

By Kaihua Ding

llms safety

arXiv AI

Sep 24

Ask Which, Not How Good: Sizing Benchmarks Scored by an LLM

The study analyzes 373,019 judgments from LLM‑scored benchmarks, decomposing variance into system, item, judge, and interaction components via generalizability theory. It finds that with a single judge, generalizability converges to a ceiling determined by the system‑by‑judge variance, which is substantially lower in pairwise preference settings, allowing one judge to suffice. The research also reveals significant biases in presentation order and highlights that many published win‑rate claims fall below the measured floor of the benchmarks.

By Atul Anand

llms benchmarks safety

arXiv AI

Sep 10

A Layered Analysis of Disagreement And Answer Quality in Multi-Agent LLM Debate

arXiv:2609.08016v1 Announce Type: new Abstract: Multi-agent debate, in which several LLMs exchange arguments before answering, is widely assumed to improve answer quality by surfacing genuine disagre...

By Chen Qian

llms agents safety

arXiv AI

Jun 16

Sycophancy as Material Failure under Pushback Loading: A Multi-Axis Characterization Across Three Loading Cases and up to Seventeen Material Charges

arXiv:2606. 16617v1 Announce Type: cross Abstract: Sycophancy in LLMs is documented across 70+ papers, but expert agreement on construct boundaries remains low (ICC=.

By Ferdinand M. Schessl

llms benchmarks

arXiv Machine Learning

Sep 22

Resist, Update, Reject: Preference Optimization Installs a Prior-Dependent Reliability Switch

The paper demonstrates that a preference‑optimization objective can learn to distinguish reliable from unreliable sources by installing a prior‑dependent reliability switch. By training on data where a source’s stated reliability is paired with its answer, the model learns to flip its response only when the stated reliability exceeds a threshold that grows with the model’s prior. Experiments on Qwen2.5‑7B‑Instruct and Llama‑3.1‑8B show that this switch generalizes to unseen reliability values and follows stated reliability over role prestige, whereas supervised imitation fails to learn it.

By Sen Yang, Yuen-Hei Yeung

llms benchmarks

arXiv AI

Jul 28

Tag Questions and the Generational Reversal of Sycophancy Across 45 Language Models

arXiv:2607. 23976v1 Announce Type: cross Abstract: Appending a two-word confirmation tag to a decision question -- "Is X the better choice?

By Tapan Parikh

llms rag benchmarks

arXiv AI

Jul 14

LLMs as a Jury: Cross-Model Consensus Can Outperform Process Reward Models for LLM Reasoning

arXiv:2607. 10139v1 Announce Type: cross Abstract: Selecting the correct answer from a pool of candidate reasoning chains is the engine of test-time scaling, yet the standard selectors each carry a cost: self-consistency inherits the errors of the single model it resamples, and trained reward models need labeled data and transfer poorly off-distribution.

By Ning Liu

llms benchmarks

arXiv AI

Oct 7

Unanimously Wrong: Certified Abstention from How Medical LLM Consensus Forms

arXiv:2610.07570v1 Announce Type: new Abstract: In clinical practice, agreement among independent experts is treated as evidence of reliability, and multi-round consensus has become a core mechanism...

By Xiaoyang Wang, Tianrui Wang, Christopher C. Yang

llms rag agents benchmarks

arXiv Machine Learning

Sep 25

Return or Revise? Learning When Revision Helps Retrieval-Augmented QA

The paper investigates when it is better to return an existing draft answer or revise it using retrieved evidence in retrieval‑augmented QA systems. By grading both the draft and its candidate revision with the same correctness judge, the authors define a paired effect called recoverability and train policies to predict it before revision. Experiments on 25,870 open‑domain questions show that a recoverability‑based scorer outperforms a draft‑correctness scorer across multiple Llama setups, improving accuracy–revision trade‑offs and closing a significant portion of the oracle gap, though it still applies harmful revisions in a substantial fraction of cases.

By Nicholas Kashani Motlagh, Tim Anderson, Jeremy Gwinnup, Grant Erdmann

llms rag

Related stories