Controlled ablation finds Popperian code-generation skill adds no separable correctness benefit over labels-only scaffold; gains track structure not content.
Play Favorites: A Statistical Method to Measure Self-Bias in
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
CoEval generates task-specific benchmarks by rotating models through teacher, student, and judge roles, then weights questions by discriminative power and judges by panel consensus to recover accurate model rankings without labels.
Anonymization in multi-agent debate reduces identity bias by equalizing self and peer weights in a Bayesian update model, quantified by the Identity Bias Coefficient.
Self-preference bias persists in rubric-based LLM evaluation even with fully objective, programmatically verifiable rubrics, and can shift subjective medical-chat scores by up to ~10 points.
Both humans and LLMs trust content more when labeled human-authored than AI-generated, with LLMs showing denser attention to labels and higher uncertainty under AI labels, mirroring human heuristic patterns.
Distilling safe refusal behavior from OpenAI o1-mini into Llama-3, Gemma-2, and Qwen3 models via response-based LoRA on multilingual jailbreak data increases jailbreak success rates on MultiJail by up to 16.6 points.
Eight LLMs exhibited massive self-preference that followed assigned identities rather than true ones, appearing in both simple word tasks and consequential evaluations of job candidates and AI technologies.
RedNote-Vibe supplies a longitudinal dataset of AI versus human lifestyle posts from 2020 to mid-2025 plus the PLAD detection framework that applies cognitive psychology signatures for improved AI-text identification.
citing papers explorer
-
Scaffold, Not Vocabulary? A Controlled, Two-Tier, Pre-Registered Study of a Popperian Code-Generation Skill
Controlled ablation finds Popperian code-generation skill adds no separable correctness benefit over labels-only scaffold; gains track structure not content.
-
CoEval: Ranking Language Models for Custom Tasks Without Labeled Data or Trustworthy Benchmarks
CoEval generates task-specific benchmarks by rotating models through teacher, student, and judge roles, then weights questions by discriminative power and judges by panel consensus to recover accurate model rankings without labels.
-
When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning
Anonymization in multi-agent debate reduces identity bias by equalizing self and peer weights in a Bayesian update model, quantified by the Identity Bias Coefficient.
-
Self-Preference Bias in Rubric-Based Evaluation of Large Language Models
Self-preference bias persists in rubric-based LLM evaluation even with fully objective, programmatically verifiable rubrics, and can shift subjective medical-chat scores by up to ~10 points.
-
Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judge
Both humans and LLMs trust content more when labeled human-authored than AI-generated, with LLMs showing denser attention to labels and higher uncertainty under AI labels, mirroring human heuristic patterns.
-
Response-Based Knowledge Distillation for Multilingual Jailbreak Prevention Unwittingly Compromises Safety
Distilling safe refusal behavior from OpenAI o1-mini into Llama-3, Gemma-2, and Qwen3 models via response-based LoRA on multilingual jailbreak data increases jailbreak success rates on MultiJail by up to 16.6 points.
-
Extreme Self-Preference in Language Models
Eight LLMs exhibited massive self-preference that followed assigned identities rather than true ones, appearing in both simple word tasks and consequential evaluations of job candidates and AI technologies.
-
RedNote-Vibe: A Dataset for Capturing Temporal Dynamics of AI-Generated Text in Lifestyle Social Media
RedNote-Vibe supplies a longitudinal dataset of AI versus human lifestyle posts from 2020 to mid-2025 plus the PLAD detection framework that applies cognitive psychology signatures for improved AI-text identification.
- Why Do Safety Guardrails Degrade Across Languages?