Abstention Is Not Threshold Compliance: Stated Scoring Rules in Open Language Models
cs.CL, cs.AI
Submitted: 2026-11-10
Updated: 2026-11-10
Terminology
Sources
- Phi-4 Technical Report
- Can a Model Catch Its Own Hallucinations for Free?: Label-Free Doubt Signals Hold Their Own Against a Labelled Dataset for Abstention
- Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning
- Confident in a Confidence Score: Investigating the Sensitivity of Confidence Scores to Supervised Fine-Tuning
- SimpleQA Verified: A Reliable Factuality Benchmark to Measure Parametric Knowledge
- The Alignment Paradox: How Post-Training Amplifies Confident Hallucinations in Language Models
- Mistral 7B
- Language Models (Mostly) Know What They Know
- Why Language Models Hallucinate
- Penalty-Framed No-Valid-Option MCQA: Analyzing LLM Abstention under Invalid Choices
- Post-Training Shifts Confidence: A Three-Stage Analysis of How SFT, RL, and OPD Shape CoT Calibration
- LLM Abstention Can Be a Prompt Artifact, in Addition to Genuine Uncertainty
- Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitation
- gpt-oss-120b & gpt-oss-20b Model Card
- Qwen2.5 Technical Report
- Qwen3 Technical Report
- 2 OLMo 2 Furious
- Olmo 3
- Are LLM Decisions Faithful to Verbal Confidence?
- Measuring short-form factuality in large language models
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data