10/ Retrieval-Augmented Fine-Tuning Combines the benefits of RAG and fine-tuning to improve a model's ability to answer questions in "open-book" in-domain settings; combining it with RAFT's CoT-style response helps to improve reasoning.
@dair_ai
-
LLMs Leak Proprietary Information Through API Logits
By
–
8/ LLMs Leak Proprietary Information Shows that it’s possible to learn a large amount of non-public information about an API-protected LLM using the logits; with a relatively small number of API queries, the approach estimates that the embedding size of gpt-3.5-turbo to be
-
DROID: Large-Scale Open-Source Robot Manipulation Dataset
By
–
9/ DROID
— DAIR.AI (@dair_ai) 24 mars 2024
An open-source, large-scale robot manipulation dataset to train and build more capable and robust robotic manipulation policies; it contains 76K demonstration trajectories, collected across 564 scenes and 86 tasks.https://t.co/8xAIyepZ0u9/ DROID An open-source, large-scale robot manipulation dataset to train and build more capable and robust robotic manipulation policies; it contains 76K demonstration trajectories, collected across 564 scenes and 86 tasks.
-
Agent-FLAN: Fine-tuning Language Models for Enhanced Agent Performance
By
–
7/ Agent-FLAN Designs data and methods to effectively fine-tune language models for agents, referred to as Agent-FLAN; this enables Llama2-7B to outperform prior best works by 3.5% across various agent evaluation datasets.
-
LLM4Decompile: Open-Source Decompilation Models Up to 33B
By
–
6/ LLM4Decompile A family of open-access decompilation LLMs ranging from 1B to 33B parameters; these models are trained on 4 billion tokens of C source code and corresponding assembly code.
-
RankPrompt: Self-Ranking Method for Better LLM Reasoning
By
–
5/ Step-by-Step Comparisons Make LLMs Better Reasoners Proposes RankPrompt, a prompting method to enable LLMs to self-rank their responses without additional resources.
-
Tool Use in Large Language Models: Overview and Applications
By
–
4/ Tool Use in LLMs Provides an overview of tool use in LLMs, including a formal definition of the tool-use paradigm, scenarios where LLMs leverage tool usage, and for which tasks this approach works well.
-
TacticAI: AI-Powered Football Tactics Assistant
By
–
3/ TacticAI
— DAIR.AI (@dair_ai) 24 mars 2024
An AI-powered assistant for football tactics that offer coaches a way to sample and explore alternative player setups for a corner kick routine and select the tactic with the highest predicted likelihood of success.https://t.co/F7oi9edMSM3/ TacticAI An AI-powered assistant for football tactics that offer coaches a way to sample and explore alternative player setups for a corner kick routine and select the tactic with the highest predicted likelihood of success.
-
Evolutionary Model Merge: Automating Foundation Model Development
By
–
2/ Evolutionary Model Merge
— DAIR.AI (@dair_ai) 24 mars 2024
An approach for automating foundation model development using evolution to combine open-source models; facilitates cross-domain merging where a Japanese Math LLM achieved SOTA performance on Japanese LLM benchmarks.https://t.co/0pFqLGg0EY2/ Evolutionary Model Merge An approach for automating foundation model development using evolution to combine open-source models; facilitates cross-domain merging where a Japanese Math LLM achieved SOTA performance on Japanese LLM benchmarks.
-
TripoSR: Fast 3D Mesh Generation from Single Images
By
–
10/ TripoSR
— DAIR.AI (@dair_ai) 10 mars 2024
A transformer-based 3D reconstruction model for fast feed-forward 3D generation; it can produce 3D mesh from a single image in under 0.5 seconds; improvement includes better data processing, model design, and training.https://t.co/5VL6kdC12410/ TripoSR A transformer-based 3D reconstruction model for fast feed-forward 3D generation; it can produce 3D mesh from a single image in under 0.5 seconds; improvement includes better data processing, model design, and training.