10/ Tree Search for Language Model Agents – proposes an inference-time tree search algorithm for LM agents to perform exploration and enable multi-step reasoning; it’s tested on interactive web environments and applied to GPT-4o to significantly improve performance.
LLMS
-
Monte Carlo Tree Search Achieves GPT-4 Level Mathematical Problem Solving
By
–
7/ Monte Carlos Tree Self-Refine – report to have achieved GPT-4 level mathematical olympiad solution using an approach that integrates LLMs with Monte Carlo Tree Search;
-
RAG Pipeline Shortcuts: How LLMs Bias Toward Context
By
–
8/ From RAG to Rich Parameters – investigates more closely how LLMs utilize external knowledge over parametric information for factual queries; finds that in a RAG pipeline, LLMs take a “shortcut” and display a strong bias towards utilizing only the context information to answer
-
PlanRAG: Iterative Plan-Then-RAG for Enhanced Decision Making
By
–
5/ PlanRAG – enhances decision making with a new RAG technique called iterative plan-then-RAG (PlanRAG); involves two steps: 1) an LM generates the plan for decision making by examining data schema and questions and 2) the retriever generates the queries for data analysis; the
-
Goldfish Loss: Mitigating Memorization in Large Language Models
By
–
6/ Mitigating Memorization in LLMs – presents a modification of the next-token prediction objective called goldfish loss to help mitigate the verbatim generation of memorized training data.
-
DeepSeek-Coder-V2 Outperforms GPT-4-Turbo on Code and Math
By
–
2/ DeepSeek-Coder-V2 – competes with closed-sourced models on code and math generation tasks; achieves 90.2% on HumanEval and 75.7% on MATH; these results are higher than GPT-4-Turbo-0409 performance according to their report; includes a 16B and 236B parameter model with 128K
-
Long-Context LLMs Performance Analysis on Retrieval and Reasoning
By
–
4/ Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More? – conducts a deep performance analysis of long-context LLMs on in-context retrieval and reasoning; they first present a benchmark with real-world tasks requiring 1M token context.
-
TextGrad: Automatic Differentiation Through LLM Textual Feedback
By
–
3/ TextGrad – a new framework for automatic differentiation through backpropagation on textual feedback provided by an LLM; this improves individual components and the natural language helps to optimize the computation graph…
-
Demo disponible avec LLaVA et GPT-4o
By
–
There is also a simple demo(LLaVA and GPT-4o) attached to the slides.
-
Multimodal LLMs: Analysis and Applications from Recent Talks
By
–
[Slides] Multimodal LLMs Recently, I gave a talk about multimodal LLMs at IndabaX Rwanda and 18-661 Introduction to Machine Learning for Engineers in Student Research Lecture at CMU. In the talk, we dissect multimodal LLMs, multimodal tasks, and a general multimodal LLM