arXiv Machine Learning
Decomposing Refusal Steering in Mixture-of-Experts Models
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Jun 4
Expert-Aware Refusal Steering
arXiv:2606. 04160v1 Announce Type: cross Abstract: Safety alignment in instruction-tuned large language models (LLMs) depends on a model's ability to reliably refuse to respond to harmful or disallowed requests.
arXiv Computation and Language
Aug 25
RASET: Router-Agnostic Safety-Critical Expert Tuning Exposes Localized Safety Enforcement Failures in Mixture-of-Experts LLMs
The paper introduces RASET, a router‑agnostic safety‑critical expert tuning framework for Mixture‑of‑Experts (MoE) large language models. RASET identifies a small subset of experts that are responsible for safety enforcement and applies parameter‑efficient tuning only to those experts, preserving the model’s intrinsic routing behavior. Experiments on five open‑weight MoE backbones show that RASET achieves a high safety‑bypass yield, outperforming existing baselines by a significant margin.
arXiv Computation and Language
Aug 24
RARE: Decoupling Representation Steering from Expert Routing in Mixture-of-Experts Language Models
The paper introduces RARE, a router‑agnostic representation engineering framework for Mixture‑of‑Experts language models. RARE projects behavioral perturbations onto the null space of the router matrix to avoid affecting routing, and corrects downstream routing drift. Experiments on six open‑weight MoE models show that RARE improves steering tasks—reducing harmfulness, increasing truthfulness, and enhancing factual editing—while preserving overall model accuracy.
arXiv Computation and Language
Sep 1
ALTSTEER: Selective Safety Steering for Moving Beyond Hard Refusals to Constructive Alternatives
arXiv:2608.30197v1 Announce Type: new Abstract: Safety alignment is essential for deploying large language models, requiring systems to prevent harmful compliance while preserving helpfulness on beni...
arXiv AI
2d ago
Mitigating LLM Over-Refusal via Dynamic Semantic Routing Calibration
arXiv:2609.25049v2 Announce Type: replace-cross Abstract: Large language models (LLMs) aligned for safety often suffer from over-refusal, incorrectly rejecting benign yet safety-related instructions....
arXiv Computation and Language
Sep 23
Mitigating LLM Over-Refusal via Dynamic Semantic Routing Calibratione
The paper addresses the problem of over‑refusal in safety‑aligned large language models, where benign instructions are incorrectly rejected. It identifies that a small set of hypersensitive safety heads in transformer attention misfire on hard‑safe prompts, causing abnormal attention entanglement and high‑entropy routing conflicts that block necessary attention to target entities. To mitigate this, the authors propose Semantic Routing Calibration (SRC), a lightweight, training‑free inference framework that dynamically suppresses these hypersensitive heads and fuses logits from dual branches to restore trustworthy reasoning while preserving intrinsic safety performance.