??Top ML Papers of the Week

The top ML Papers of the Week (Oct 16 - Oct 22):

1). Llemma - an LLM for mathematics which is based on continued pretraining from Code Llama on the Proof-Pile-2 dataset; the dataset involves scientific paper, web data containing mathematics, and mathematical code; Llemma outperforms open base models and the unreleased Minerva on the MATH benchmark; the model is released, including dataset and code to replicate experiments. (paper | tweet)

2). LLMs for Software Engineering - a comprehensive survey of LLMs for software engineering, including open research and technical challenges. (paper | tweet)

3). Self-RAG - presents a new retrieval-augmented framework that enhances an LM’s quality and factuality through retrieval and self-reflection; trains an LM that adaptively retrieves passages on demand, and generates and reflects on the passages and its own generations using special reflection tokens; it significantly outperforms SoTA LLMs (ChatGPT and retrieval-augmented Llama2-Chat) on open-domain QA, reasoning, and fact verification tasks, including factuality improvements. (paper | tweet)

4). Retrieval-Augmentation for Long-form Question Answering - explores retrieval-augmented language models on long-form question answering; finds that retrieval is an important component but evidence documents should be carefully added to the LLM; finds that attribution error happens more frequently when retrieved documents lack sufficient information/evidence for answering the question. (paper | tweet)

5). GenBench - presents a framework for characterizing and understanding generalization research in NLP; involves a meta-analysis of 543 papers and a set of tools to explore and better understand generalization studies. (paper | tweet)

Sponsor message

DAIR.AI presents a new cohort-based course, Prompt Engineering for LLMs, that teaches how to effectively use the latest prompt engineering techniques and tools to improve the capabilities, performance, and reliability of LLMs. Enroll here.

6). A Study of LLM-Generated Self-Explanations - assesses an LLM's capability to self-generate feature attribution explanations; self-explanation is useful to improve performance and truthfulness in LLMs; this capability can be used together with chain-of-thought prompting. (paper | tweet)

7). OpenAgents - an open platform for using and hosting language agents in the wild; includes three agents, including a Data Agent for data analysis, a Plugins Agent with 200+ daily API tools, and a Web Agent for autonomous web browsing. (paper | tweet)

8) Eliciting Human Preferences with LLMs - uses language models to guide the task specification process and a learning framework to help models elicit and infer intended behavior through free-form, language-based interaction with users; shows that by generating open-ended questions, the system generates responses that are more informative than user-written prompts. (paper | tweet)

9). AutoMix - an approach to route queries to LLMs based on the correctness of smaller language models (done via few-shot self-verification); a meta-verifier is introduced to check the verifier's output (typically a smaller model) and route the query to a larger language model if needed. Experiments using LLAMA2-13/70B, on five context-grounded reasoning datasets demonstrate that AutoMix surpasses established baselines, improving the incremental benefit per cost by up to 89%. (paper | tweet)

10). Video Language Planning - enables synthesizing complex long-horizon video plans across robotics domains; the proposed algorithm involves a tree search procedure that trains vision-language models to serve as policies and value functions, and text-to-video models as dynamic models. (paper | tweet)