BAIR – Berkeley Artificial Intelligence Research AI news
View books and computing supplies on the AI industry from Amazon
-
RL without TD learning
In this post, I’ll introduce a reinforcement learning (RL) algorithm based on an “alternative” paradigm: divide and conquer. Unlike traditional methods, this algorithm is not based on temporal difference (TD) learning (which has scalability challenges), and scales well to long-horizon
... -
What exactly does word2vec learn?
What exactly does
...word2veclearn, and how? Answering this question amounts to understanding representation learning in a minimal yet interesting language modeling task. Despite the fact thatword2vecis a well-known precursor to modern language models, for many years, -
Whole-Body Conditioned Egocentric Video Prediction
.modal { display: none; position: fixed; z-index: 9999; padding-top: 50px; left: 0; top: 0; width: 100%; height: 100%; overflow: auto; background-color: rgba(0,0,0,0.9); } .modal-content { margin: auto; display: block; max-width: 90%; max-height: 90%; } .close { position: absolute; top: 15px; right: 35px; color: #f1f1f1; font-size: 40px; font-weight:... -
Defending against Prompt Injection with Structured Queries (StruQ) and Preference Optimization (SecAlign)
Recent advances in Large Language Models (LLMs) enable exciting LLM-integrated applications. However, as LLMs have improved, so have the attacks against them. Prompt injection attack is listed as the #1 threat by OWASP to LLM-integrated
... -
Repurposing Protein Folding Models for Generation with Latent Diffusion
...
PLAID is a multimodal generative model that simultaneously generates protein 1D sequence and 3D structure, by learning the latent space of protein folding -
Scaling Up Reinforcement Learning for Traffic Smoothing: A 100-AV Highway Deployment
Training Diffusion Models with Reinforcement Learning -
Virtual Personas for Language Models via an Anthology of Backstories
--
-->...
We introduce Anthology, a method for conditioning LLMs to representative, consistent, and diverse virtual personas by generating and utilizing naturalistic backstories with rich details of individual values and experience. -
Linguistic Bias in ChatGPT: Language Models Reinforce Dialect Discrimination
Sample language model responses to different varieties of English and native speaker reactions.ChatGPT does amazingly well at communicating with people in English. But whose English?
-
How to Evaluate Jailbreak Methods: A Case Study with the StrongREJECT Benchmark
When we began studying jailbreak evaluations, we found a fascinating paper claiming that you could jailbreak frontier LLMs simply by translating forbidden prompts into obscure languages. Excited by this result, we attempted to reproduce it and found something unexpected.
The paper in question claimed an impressive 43% success rate in jailbreaking GPT-4 by translating forbidden
... -
Are We Ready for Multi-Image Reasoning? Launching VHs: The Visual Haystacks Benchmark!
Humans excel at processing vast arrays of visual information, a skill that is crucial for achieving artificial general intelligence (AGI). Over the decades, AI researchers have developed Visual Question Answering (VQA) systems to interpret scenes within single images and answer related questions. While recent advancements in foundation models have significantly closed the gap between human
...