Beyond Hidden-Layer Manipulation: Semantically-Aware Logit Interventions for Debiasing LLMs
We proposed Static and Dynamic -- two zero-shot logits-layer debiasing methods. Dynamic reduces bias by up to 70% with minimal fluency loss. Logits intervention outperforms hidden-layer approaches. We show semantic-aware logits intervention is stable and effective for debiasing aligned LLMs.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
A Disentangling Invertible Interpretation Network for Explaining Latent Representations
Neural networks have greatly boosted performance in computer vision by learning powerful representations of input data. The drawback of end-to-end training for maximal overall performance are black-box models whose hidde…
Image GenerationImage ManipulationD2PPO: Diffusion Policy Policy Optimization with Dispersive Loss
Diffusion policies excel at robotic manipulation by naturally modeling multimodal action distributions in high-dimensional spaces. Nevertheless, diffusion policies suffer from diffusion representation collapse: semantica…
Jailbreaking Embodied LLMs via Action-level Manipulation
Embodied Large Language Models (LLMs) enable AI agents to interact with the physical world through natural language instructions and actions. However, beyond the language-level risks inherent to LLMs themselves, embodied…
Spatiotemporal Hidden-State Dynamics as a Signature of Internal Reasoning in Large Language Models
Large reasoning models (LRMs) generate extended solutions, yet it remains unclear whether these traces reflect substantive internal computation or merely verbosity and overthinking. Although recent hidden-state analyses …
HANDO: Hierarchical Autonomous Navigation and Dexterous Omni-loco-manipulation
Seamless loco-manipulation in unstructured environments requires robots to leverage autonomous exploration alongside whole-body control for physical interaction. In this work, we introduce HANDO (Hierarchical Autonomous …