Search Results for author:
Found 7 papers, 4 papers with code
SAE-V: Interpreting Multimodal Models for Enhanced Alignment
In text-only LLMs, Sparse Autoencoders (SAEs) have gained attention for their ability to interpret latent representations.
Stream Aligner: Efficient Sentence-Level Alignment via Distribution Induction
The rapid advancement of large language models (LLMs) has led to significant improvements in their capabilities, but also to increased concerns about their alignment with human values and intentions.
Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback
In this work, we make the first attempt to fine-tune all-modality models (i. e. input and output with any modality, also named any-to-any models) using human preference data across all modalities (including text, image, audio, and video), ensuring its behavior aligns with human intentions.
Language Models Resist Alignment: Evidence From Data Compression
Empirically, we demonstrate the elasticity of post-alignment models, i. e., the tendency to revert to the behavior distribution formed during the pre-training phase upon further fine-tuning.
Aligner: Efficient Alignment by Learning to Correct
However, the tension between the complexity of current alignment methods and the need for rapid iteration in deployment scenarios necessitates the development of a model-agnostic alignment approach that can operate under these constraints.
AI Alignment: A Comprehensive Survey
The former aims to make AI systems aligned via alignment training, while the latter aims to gain evidence about the systems' alignment and govern them appropriately to avoid exacerbating misalignment risks.
Deep learning-based rapid generation of broadly reactive antibodies against SARS-CoV-2 and its Omicron variant
The COVID-19 pandemic has been ongoing for nearly two and half years, and new variants of concern (VOCs) of SARS-CoV-2 continue to emerge, which urges the development of broadly neutralizing antibodies.
