- JudgeLM: Fine-tuned Large Language Models are Scalable Judges - [Arxiv] [QA]
- Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time - [Arxiv] [QA]
- HyperFields: Towards Zero-Shot Generation of NeRFs from Text - [Arxiv] [QA]
- Controlled Decoding from Language Models - [Arxiv] [QA]
- LLM-FP4: 4-Bit Floating-Point Quantized Transformers - [Arxiv] [QA]
- LightSpeed: Light and Fast Neural Light Fields on Mobile Devices - [Arxiv] [QA]
- TD-MPC2: Scalable, Robust World Models for Continuous Control - [Arxiv] [QA]
- CommonCanvas: An Open Diffusion Model Trained with Creative-Commons Images - [Arxiv] [QA]
- DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior - [Arxiv] [QA]
- QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models - [Arxiv] [QA]
- Detecting Pretraining Data from Large Language Models - [Arxiv] [QA]
- ConvNets Match Vision Transformers at Scale - [Arxiv] [QA]
- A Picture is Worth a Thousand Words: Principled Recaptioning Improves Image Generation - [Arxiv] [QA]
- An Early Evaluation of GPT-4V(ision) - [Arxiv] [QA]
- CLEX: Continuous Length Extrapolation for Large Language Models - [Arxiv] [QA]
- TiC-CLIP: Continual Training of CLIP Models - [Arxiv] [QA]
- Woodpecker: Hallucination Correction for Multimodal Large Language Models - [Arxiv] [QA]
- What's Left? Concept Grounding with Logic-Enhanced Foundation Models - [Arxiv] [QA]
- Dissecting In-Context Learning of Translations in GPTs - [Arxiv] [QA]
- In-Context Learning Creates Task Vectors - [Arxiv] [QA]
- KITAB: Evaluating LLMs on Constraint Satisfaction for Information Retrieval - [Arxiv] [QA]
- TRAMS: Training-free Memory Selection for Long-range Language Modeling - [Arxiv] [QA]
- Moral Foundations of Large Language Models - [Arxiv] [QA]
- SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding - [Arxiv] [QA]
- RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptions - [Arxiv] [QA]
- Papers for 2023
- Papers for 2022
- Papers for 2021
- Papers for 2020
- Papers for 2019
- Papers for 2018
- Papers for 2017
- Papers for 2016
- Papers for 2015
- Papers for 2014
- Papers for 2013
- Papers for 2012
- Papers for 2010
- Papers for 2009
This project is made possible through the generous support of Anthropic, who provided free access to the Claude-2.0
API.