REVIEW 36 cited by
Automatic Prompt Optimization with "Gradient Descent" and Beam Search
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Large Language Models (LLMs) have shown impressive performance as general purpose agents, but their abilities remain highly dependent on prompts which are hand written with onerous trial-and-error effort. We propose a simple and nonparametric solution to this problem, Automatic Prompt Optimization (APO), which is inspired by numerical gradient descent to automatically improve prompts, assuming access to training data and an LLM API. The algorithm uses minibatches of data to form natural language "gradients" that criticize the current prompt. The gradients are then "propagated" into the prompt by editing the prompt in the opposite semantic direction of the gradient. These gradient descent steps are guided by a beam search and bandit selection procedure which significantly improves algorithmic efficiency. Preliminary results across three benchmark NLP tasks and the novel problem of LLM jailbreak detection suggest that Automatic Prompt Optimization can outperform prior prompt editing techniques and improve an initial prompt's performance by up to 31%, by using data to rewrite vague task descriptions into more precise annotation instructions.
Forward citations
Cited by 36 Pith papers
-
DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines
DSPy compiles short declarative programs into LM pipelines that self-optimize and outperform both standard few-shot prompting and expert-written chains on math, retrieval, and QA tasks.
-
Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution
Promptbreeder evolves both task prompts and the mutation prompts that improve them using LLMs, outperforming Chain-of-Thought and Plan-and-Solve on arithmetic and commonsense reasoning benchmarks.
-
Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories
A dedicated harness-editor policy trained with RL on the realized outcomes of executable patches raises frozen-agent success by 9.3 points across WebShop, ALFWorld, and DBBench.
-
Meta-Harness: End-to-End Optimization of Model Harnesses
Meta-Harness discovers improved harness code for LLMs via agentic search over prior execution traces, yielding 7.7-point gains on text classification with 4x fewer tokens and 4.7-point gains on math reasoning across h...
-
CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training
CapTrack shows post-training causes drift beyond facts, with instruction fine-tuning producing stronger behavioral changes than preference optimization across model families.
-
PIAST: Rapid Prompting with In-context Augmentation for Scarce Training data
PIAST iteratively optimizes few-shot examples in prompts via Monte Carlo Shapley value estimation, outperforming prior automatic prompting methods and setting new SOTA on classification, simplification, and GSM8K with...
-
ToolGrad: Efficient Tool-use Dataset Generation with Textual "Gradients"
ToolGrad inverts the standard tool-use dataset synthesis process by constructing valid tool chains first with textual gradients, producing a high-quality 500-example dataset with near-perfect validity and superior mod...
-
EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers
EvoPrompt uses LLMs to run evolutionary operators on populations of prompts, outperforming human-engineered prompts by up to 25% on BIG-Bench Hard tasks across 31 datasets.
-
Large Language Models as Optimizers
Large language models can optimize by being prompted with histories of past solutions and scores to propose better ones, producing prompts that raise accuracy up to 8% on GSM8K and 50% on Big-Bench Hard over human-des...
-
Reflexion: Language Agents with Verbal Reinforcement Learning
Reflexion lets LLM agents improve via stored verbal reflections on task feedback, reaching 91% pass@1 on HumanEval and outperforming prior GPT-4 results.
-
Asking Questions the Right Way: A Multi-Agent Conversational System for Prompt Formulation in Complex Task Resolution
An eight-agent question-asking system that front-loads intent clarification produced more complete prompts, higher-rated outputs, and single-turn task completion in a four-person pilot, with unstable effect sizes.
-
MemoHarness: Agent Harnesses That Learn from Experience
MemoHarness learns from past executions to adapt an LLM agent's control layer per test case, improving task success over fixed harnesses on shell, code, and reasoning benchmarks.
-
Heuristic Learning for Active Flow Control Using Coding Agents
Under equal simulation budgets, coding-agent search finds explicit flow controllers that match or beat the best DRL baseline on 10 of 13 active-flow-control benchmarks.
-
BT-APE: A Computationally Light Backtracking Approach to Automatic Prompt Engineering for Requirements Classification
BT-APE automates prompt engineering for requirements classification using backtracking search and dynamic examples, matching PE2 accuracy while using 72% fewer tokens and 66% less time than that baseline.
-
MetaPS: Adaptive Programmatic Strategy Selection for Market Agents
MetaPS trains models via simulation rollouts to select from programmatic strategy libraries for market agents, yielding better performance than fixed or direct LLM baselines across model sizes.
-
BCL: Bayesian In-Context Learning Framework for Information Extraction
BCL introduces a particle-filtering Bayesian update framework to systematically refine label representations in in-context learning for information extraction, claiming consistent gains over prior methods.
-
Prompt Optimization Is a Coin Flip: Diagnosing When It Helps in Compound AI Systems
Prompt optimization in compound AI systems is statistically indistinguishable from random chance except when tasks have exploitable output structure; a two-stage diagnostic predicts success.
-
Prompt Optimization Is a Coin Flip: Diagnosing When It Helps in Compound AI Systems
End-to-end prompt optimization in compound AI systems is no better than chance unless the task has exploitable output structure the model can produce but does not default to.
-
MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural Networks
MASPOB combines a GNN surrogate, LinUCB-style uncertainty, and coordinate ascent to optimize prompts in fixed-topology multi-agent LLM systems, beating AFlow and MIPRO on average across six benchmarks.
-
Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search
Gome reaches 35.1% any-medal rate on MLE-Bench by mapping reasoning to gradient-based updates, outperforming tree search once models are sufficiently capable.
-
Visual Persuasion: What Influences Decisions of Vision-Language Models?
Naturalistic edits to background, lighting, and staging systematically shift the choices of frontier vision-language models, and a new optimization-plus-interpretation framework (CVPO) discovers and explains the visua...
-
Knowing How to Edit: Reliable Evaluation Signals for Diagnosing and Optimizing Prompts at Query Level
An execution-free evaluator that predicts prompt-quality metrics guides per-query prompt rewriting, but the reported consistent gains are not supported by the paper's own tables.
-
CaPulse: Detecting Anomalies by Tuning in to the Causal Rhythms of Time Series
A single-pass instruction rewriter trained on MCTS-synthesized preference data improves LLM alignment win rates over the BPO baseline.
-
ToolGrad: Efficient Tool-use Dataset Generation with Textual "Gradients"
ToolGrad inverts tool-use dataset generation: build valid tool-call chains first, synthesize queries second, yielding lower cost and near-100% pass rates.
-
Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting
LLMs are highly sensitive to prompt formatting in few-shot settings, with accuracy varying by up to 76 points across formats; FormatSpread samples formats to report performance intervals without model weights.
-
LLM4MTLs: Automated Generation and Empirical Evaluation of Model Transformation Languages
Few-shot prompting improves syntactic validity of LLM-generated code across ATL, ETL, QVTo, and Reactions, but semantic correctness gains remain uneven and language-dependent.
-
Towards Fast Domain Adaptation and Fine-Grained User Simulation for Evaluating Conversational Recommender Systems
AdaptSim is an adaptive user simulator for CRS evaluation that combines automatic prompt generation, open actions, controlled text generation, and BFS-based pairwise comparison to produce realistic dialogues and asses...
-
Marginal Advantage Accumulation for Memory-Driven Agent Self-Evolution
MAA formalizes alignability and comparability conditions and uses differential signals, EMA accumulation, and semantic identity merging to enable cross-batch operation-level evidence accumulation, outperforming batch-...
-
Embedding-perturbed Exploration Preference Optimization for Flow Models
E²PO uses embedding-level perturbations to maintain intra-group variance and discriminative signal in RL-based preference optimization for generative flow models.
-
Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering
LLM agent progress depends on externalizing cognitive functions into memory, skills, protocols, and harness engineering that coordinates them reliably.
-
LAD-VF: LLM-Automatic Differentiation Enables Fine-Tuning-Free Robot Planning from Formal Methods Feedback
LAD-VF iteratively rewrites LLM prompts using formal verification failure rates as a loss, lifting safety-spec compliance from about 60% to over 90% without model fine-tuning.
-
TAHOE: Text-to-SQL with Automated Hint Optimization from Experience
TAHOE builds a Hint Bank from error traces to raise Text-to-SQL pass rates on Spider 2.0-Snow from 61.95% to 79.42% for GPT-5.5 without parameter updates.
-
Building Customer Support AI Agents at 100M-User Scale: An Evaluation-Driven Framework
An evaluation-driven framework for customer support AI agents at Nubank integrates context engineering, LLM judges, and A/B testing to deliver up to 37pp NPS gains and strong offline-online correlation across five pro...
-
A Reproducibility Analysis of PO4ISR: Diagnosing and Mitigating Semantic Drift in LLM-Based Session Recommendation
Reproducibility study diagnoses semantic drift in PO4ISR and introduces PO4ISR++ with reflexive prompting that restores performance with gains up to 54% on Games and 96% on Bundle.
-
WST: Weak-to-Strong Knowledge Transfer via Reinforcement Learning
WST uses RL to train a small model to generate instructions that boost a larger model's benchmark performance, with reported gains of 98% on MATH-500 and 134% on HH-RLHF.
-
A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence
The paper delivers the first systematic review of self-evolving agents, structured around what components evolve, when adaptation occurs, and how it is implemented.
Discussion (0). Sign in to comment.