paper-with-me

Papers

Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programs

2024-06-17 · Krista Opsahl-Ong, Michael J Ryan, Josh Purtell, David Broman, Christopher Potts, Matei Zaharia, Omar Khattab

Language Model Programs, i.e. sophisticated pipelines of modular language model (LM) calls, are increasingly advancing NLP tasks, but they require crafting prompts that are jointly effective for all modules. We study prompt optimization for LM programs, i.e. how to update these prompts to maximize a downstream metric without access to module-level labels or gradients. To make this tractable, we factorize our problem into optimizing the free-form instructions and few-shot demonstrations of every module and introduce several strategies to craft task-grounded instructions and navigate credit assignment across modules. Our strategies include (i) program- and data-aware techniques for proposing effective instructions, (ii) a stochastic mini-batch evaluation function for learning a surrogate model of our objective, and (iii) a meta-optimization procedure in which we refine how LMs construct proposals over time. Using these insights we develop MIPRO, a novel algorithm for optimizing LM programs. MIPRO outperforms baseline optimizers on five of seven diverse multi-stage LM programs using a best-in-class open-source model (Llama-3-8B), by as high as 13% accuracy. We have released our new optimizers and benchmark in DSPy at http://dspy.ai

📄 PDF Abstract BibTeX arXiv:2406.11695

Code (2)

stanfordnlp/dspy 공식 구현 pytorch
evoagentx/evoagentx

Tasks

Language ModelingLanguage ModellingNavigate

Similar Papers 제목 키워드 기반

GoalNet: Inferring Conjunctive Goal Predicates from Human Plan Demonstrations for Robot Instruction Following

2022-05-14 · Shreya Sharma, Jigyasa Gupta, Shreshth Tuli, Rohan Paul 외

Our goal is to enable a robot to learn how to sequence its actions to perform tasks specified as natural language instructions, given successful demonstrations from a human partner. The ability to plan high-level tasks c…

Decision MakingInstruction Following

Improving Generalization of Language-Conditioned Robot Manipulation

2025-08-04 · Chenglin Cui, Chaoran Zhu, Changjae Oh, Andrea Cavallaro arxiv

The control of robots for manipulation tasks generally relies on visual input. Recent advances in vision-language models (VLMs) enable the use of natural language instructions to condition visual input and control robots…

Robot Manipulation

Ask Your Humans: Using Human Instructions to Improve Generalization in Reinforcement Learning

2020-11-01 · ICLR 2021 1 · Valerie Chen, Abhinav Gupta, Kenneth Marino

Complex, multi-task problems have proven to be difficult to solve efficiently in a sparse-reward reinforcement learning setting. In order to be sample efficient, multi-task learning requires reuse and sharing of low-leve…

Multi-Task Learningreinforcement-learningReinforcement Learning (RL)

NNetscape Navigator: Complex Demonstrations for Web Agents Without a Demonstrator

2024-10-03 · Shikhar Murty, Dzmitry Bahdanau, Christopher D. Manning

We introduce NNetscape Navigator (NNetnav), a method for training web agents entirely through synthetic demonstrations. These demonstrations are collected by first interacting with a browser to generate trajectory rollou…

Language ModelingLanguage Modelling

Explaining Multi-stage Tasks by Learning Temporal Logic Formulas from Suboptimal Demonstrations

2020-06-03 · Glen Chou, Necmiye Ozay, Dmitry Berenson

We present a method for learning multi-stage tasks from demonstrations by learning the logical structure and atomic propositions of a consistent linear temporal logic (LTL) formula. The learner is given successful but po…