arXiv:2608. 11947v1 Announce Type: cross Abstract: Multiple-choice benchmarks are widely used to evaluate large language models, but MCQ scores conflate knowledge with sensitivity to option order, which makes them unreliable measures of model knowledge.
arXiv:2605.24981v2 Announce Type: replace Abstract: Choosing a Large Language Model (LLM) for a given task requires comparing many strong candidates, yet standard evaluation relies on costly annotati...
The paper introduces a bias depth score to differentiate between stable model preferences (Deep biases) and prompt‑dependent responses (Shallow biases) in large language models. By analyzing 4,442 opinion prompts across four models, it finds that only about a quarter of concentrated preferences persist after scenario reframing, indicating that most are shallow. The study shows Deep biases are more often inherited from pretraining and harder to remove through fine‑tuning or prompt‑based debiasing, highlighting the need to distinguish learned biases from prompt artifacts.
arXiv:2606.16011v2 Announce Type: replace Abstract: Standard accuracy benchmarks evaluate whether large language models (LLMs) reach correct answers. However, they do not test whether models maintain...
arXiv:2610.35826v1 Announce Type: new Abstract: Selection bias in large language models (LLMs) can persist despite instructions to ignore presentation order or prompts that elicit reasoning. Recent e...
arXiv:2604. 04944v2 Announce Type: replace-cross Abstract: Multiple-choice questions (MCQs) are widely used to evaluate large language models (LLMs).
arXiv:2607. 14109v1 Announce Type: cross Abstract: Probing the capabilities of Large Language Models (LLMs) and building robust solutions for Multiple-Choice Question Answering (MCQA) remain central challenges in natural language understanding.
The paper introduces Length‑Conditioned Majority Voting (LCMV), a test‑time training method that enhances the standard Majority Voting (MV) technique for Large Language Models (LLMs). LCMV incorporates an empirical bias that longer Chain‑of‑Thoughts (CoTs) are less likely to be correct by modeling a length‑conditioned prior with a logistic function. This prior is adapted on unlabeled test data without fine‑tuning the LLM, and the resulting weighted votes consistently outperform standard MV and other length‑ or confidence‑based aggregation strategies across multiple reasoning benchmarks.
The paper demonstrates that multiple‑choice visual question answering (MC‑VQA) benchmarks are unreliable because model performance is highly sensitive to semantically neutral prompt formatting choices—such as option ID sets, delimiters, and separators—despite protocols that mitigate option‑order effects. Across seven multimodal large language models and five datasets, the authors observed frequent rank reversals when systematically varying 48 equivalent prompt formats, attributing the instability to tokenizer‑induced token fusion or removal and to how option ID sets influence attention patterns. Consequently, MC‑VQA rankings correlate weakly with open‑ended evaluation, revealing that MC‑VQA reflects option‑selection dynamics as well as multimodal reasoning.
arXiv:2609.35879v1 Announce Type: cross Abstract: Benchmarks for large language models (LLMs) typically evaluate the accuracy of answers against fixed reference labels. But a central step in many com...
arXiv:2508. 11847v4 Announce Type: replace-cross Abstract: We propose a method for evaluating the robustness of widely used LLM ranking systems -- variants of a Bradley--Terry model -- to dropping a worst-case very small fraction of preference data.
arXiv:2607.14109v2 Announce Type: replace Abstract: Probing the capabilities of Large Language Models (LLMs) and building robust solutions for Multiple-Choice Question Answering (MCQA) remain central...