VoirDire: Counterfactual Calibration for Reliable LLM-as-a-Judge Aggregation
cs.LG
Submitted: 2026-10-31
Updated: 2026-10-31
Terminology
Sources
- Yi: Open Foundation Models by 01.AI
- Qwen2.5 Technical Report
- Qwen3 Technical Report
- Dependence-Aware Label Aggregation for LLM-as-a-Judge via Ising Models
- Gemma 2: Improving Open Language Models at a Practical Size
- Gemma 3 Technical Report
- RewardBench 2: Advancing Reward Model Evaluation
- The Llama 3 Herd of Models
- Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
- Mistral 7B
- Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
- HelpSteer2: Open-source dataset for training top-performing reward models
- CARE: Confounder-Aware Aggregation for Reliable LLM Evaluation
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis