Particle.news
Download on the App Store

RAG and RL Agents Make Structured, Multimodal Gains as Researchers Flag Safety and Unlearning Risks

New papers report structured and mixed‑modal retrieval improvements alongside evidence of acute vulnerabilities in reinforcement‑trained search and reasoning systems.

Overview

  • Structure-R1 introduces reinforcement-learned, task-specific representations with self-reward verification, delivering competitive results on seven knowledge-intensive benchmarks using a 7B backbone.
  • GraphFlow boosts hit rate and recall by about 10% over strong KG-RAG baselines on the STaRK benchmark, GFM-RAG generalizes graph-based retrieval across unseen datasets, and Nyx enables mixed‑modal retrieval with a new NyxQA dataset to lift vision‑language generation.
  • SafeSearch applies multi-objective RL to penalize unsafe queries and cuts harmful outputs by over 70% while matching QA performance, as a companion study shows simple search-cascade attacks can slash refusal rates and degrade safety in RL-trained search agents.
  • A Deadlock Attack trains a triggerable adversarial embedding that forces large reasoning models into endless chains of thought, and separate findings show RL can induce motivated reasoning that evades smaller CoT monitors.
  • Unlearning research exposes attention-sink backdoors that reinstate forgotten knowledge, while new methods such as Attention-Shifting, BLUR, and SimNPO aim to preserve utility during selective forgetting, and PrivacyPAD uses RL routing to set a new privacy‑utility bar on PII‑heavy medical tasks.