🥇Top ML Papers of the Week
The top ML Papers of the Week (April 10 - April 16)
1). Zip-NeRF - combines mip-NeRF 360 and grid-based models to improve NeRFs that train 22x faster than mip-NeRF 360. (paper | tweet)
2). LLMs as Generative Agents - proposes an architecture that extends LLMs to build agents that enable simulations of human-like behavior; these capabilities are possible by storing a complete record of an agent's experiences, synthesizing memories over time into higher-level reflections, and retrieving them dynamically to plan behavior. (paper | tweet)
3). Scientific Research Capabilities of LLMs - presents an agent that combines LLMs for autonomous design, planning, and execution of scientific experiments; shows emergent scientific research capabilities, including the successful performance of catalyzed cross-coupling reactions. (paper | tweet)
4). Automatic Gradient Descent - derives optimization algorithms that explicitly leverage neural architecture; it proposes a first-order optimizer without hyperparameters that trains CNNs at ImageNet scale. (paper | tweet)
5). ChemCrow - presents an LLM chemistry agent that performs tasks across synthesis, drug discovery, and materials design; it integrates 13 expert-design tools to augment LLM performance in chemistry and demonstrate effectiveness in automating chemical tasks. (paper | tweet)
6). A Survey of ChatGPT and GPT-4 (paper | tweet)
7). OpenAGI - an open-source research platform to facilitate the development and evaluation of LLMs in solving complex, multi-step tasks through manipulating various domain expert models. (paper | tweet)
8). AGIEval - a new benchmark to assess foundational models in the context of human-centric standardized exams, including college entrance exams, law school admission tests, and math competitions, among others. (paper | tweet)
9). Teaching LLMs to Self-Debug - proposes an approach that teaches LLMs to debug their predicted program via few-shot demonstrations; this allows a model to identify its mistakes by explaining generated code in natural language; achieves SoTA on several code generation tasks like text-to-SQL generation. (paper | tweet)
10). Segment Everything Everywhere All at Once - a promptable, interactive model for various segmentation tasks that yields competitive performance on open-vocabulary and interactive segmentation benchmarks. (paper | tweet)
Please note that Substack is having issues embedding tweets so we provided a tweet link instead. Thanks for all your support.