META-IR: Meta-Introspective Reasoning for Adaptive Safety Alignment in LLM
cs.AI
Submitted: 2026-11-06
Updated: 2026-11-06
Code: https://github.com/Miraj-Rahman-AI/Meta-IR
Terminology
Sources
- Beyond Fixed and Dynamic Prompts: Embedded Jailbreak Templates for Advancing LLM Security
- Adversarial Preference Learning for Robust LLM Alignment
- STAIR: Improving Safety Alignment with Introspective Reasoning
- HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
- Self-rewarding correction for mathematical reasoning
- HarmMetric Eval: Benchmarking Metrics and Judges for LLM Harmfulness Assessment
- A Critical Look at Meta-evaluating Summarisation Evaluation Metrics
- am-ELO: A Stable Framework for Arena-based LLM Evaluation
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents