ShallowDefend puts an explicit safety decision before multi-turn replies
Assess, Decide, Respond is trained with reinforcement learning and led multi-turn safety on 14 benchmarks across two model families.
Stories per day · 30 days
2 RSS
Relateddpo1jailbreak-defense1llm-judge1llm-safety1manaca1multi-turn1over-refusal1portuguese1reinforcement-learning1sft1
5 stories, newest first · one card per story
Stories tagged #safety-alignment
Manacá-1B-Instruct-2: Reproducible Safety Alignment for a Native Brazilian-Portuguese LLM. Negative DPO Results, a Controlled Composition Scan, and a Silent Failure Mode of the LLM Judge
Manacá-1B-Instruct-2 gained safety from iterated safety finetuning after three failed DPO runs, beating four Portuguese baselines.
Controlled Decoding Attacks on Black-Box LLMs
A text-only controlled-decoding method jailbreaks black-box LLMs by reconstructing and steering token distributions at a few positions.
Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal
A self-distillation safety framework tunes narrow-boundary refusals in Qwen3-8B, raising target-domain refusal to 84.75% while cutting over-refusal from 15.20% to 5.20%.
ShieldVLA: Feasibility-Aware Safety Alignment for Vision-Language-Action Models
ShieldVLA, a safety-aligned fine-tuning framework for VLA models, improves safety guarantees via Hamilton-Jacobi reachability.