mlsanityData to 14 Oct 2026

TextGrad: Automatic "Differentiation" via Text

arXiv 2406.07496 ↗First posted 11 Jun 2024cs.CL cs.AI cs.LGcited by 273

Not on the map. Why?

Cited by 120 in LLM agents and tool use, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

AI is undergoing a paradigm shift, with breakthroughs achieved by systems orchestrating multiple large language models (LLMs) and other complex components. As a result, developing principled and automated optimization methods for compound AI systems is one of the most important new challenges. Neural networks faced a similar challenge in its early days until backpropagation and automatic differentiation transformed the field by making optimization turn-key. Inspired by this, we introduce TextGrad, a powerful framework performing automatic ``differentiation'' via text. TextGrad backpropagates textual feedback provided by LLMs to improve individual components of a compound AI system. In our framework, LLMs provide rich, general, natural language suggestions to optimize variables in computation graphs, ranging from code snippets to molecular structures. TextGrad follows PyTorch's syntax and abstraction and is flexible and easy-to-use. It works out-of-the-box for a variety of tasks, where the users only provide the objective function without tuning components or prompts of the framework. We showcase TextGrad's effectiveness and generality across a diverse range of applications, from question answering and molecule optimization to radiotherapy treatment planning. Without modifying the framework, TextGrad improves the zero-shot accuracy of GPT-4o in Google-Proof Question Answering from 51% to 55%, yields 20% relative performance gain in optimizing LeetCode-Hard coding problem solutions, improves prompts for reasoning, designs new druglike small molecules with desirable in silico binding, and designs radiation oncology treatment plans with high specificity. TextGrad lays a foundation to accelerate the development of the next-generation of AI systems.

Key facts. Cited by 273 papers in the map (data through 14 Oct 2026). Builds on 55 of its 107 references that are in the map. First posted on arXiv 11 Jun 2024.

Builds on

55 of 107 references.

References · 55

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2022Training language models to follow instructions with human feedbackcited by 15,971
  5. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  6. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  7. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  8. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  9. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  10. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  11. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  12. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
38 more
  1. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  2. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  3. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,915
  4. 2023Let's Verify Step by Stepcited by 3,789
  5. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  6. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  7. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  8. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,224
  9. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  10. 2020Learning to summarize from human feedbackcited by 2,588
  11. 2021Convex optimizationcited by 2,515
  12. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,234
  13. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,998
  14. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,807
  15. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineeringcited by 1,737
  16. 2022Visual Prompt Tuningcited by 1,681
  17. 2022Competition-Level Code Generation with AlphaCodecited by 1,462
  18. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  19. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,045
  20. 2023Teaching Large Language Models to Self-Debugcited by 988
  21. 2023ExpeL: LLM Agents Are Experiential Learnerscited by 876
  22. 2023Large Language Models as Optimizerscited by 864
  23. 2023AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedbackcited by 748
  24. 2024Self-Rewarding Language Modelscited by 681
  25. 2022Large Language Models Can Self-Improvecited by 664
  26. 2023Automatic Prompt Optimization with "Gradient Descent" and Beam Searchcited by 653
  27. 2023Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicinecited by 329
  28. 2022Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLPcited by 318
  29. 2014Caffe: Convolutional Architecture for Fast Feature Embeddingcited by 288
  30. 2023Self-Evaluation Guided Beam Search for Reasoningcited by 196
  31. 2022Re3: Generating Longer Stories With Recursive Reprompting and Revisioncited by 193
  32. 2021KnowPrompt: Knowledge-aware Prompt-tuning with Synergistic Optimization for Relation Extractioncited by 180
  33. 2024Large Language Models to Enhance Bayesian Optimizationcited by 135
  34. 2023Hypothesis Search: Inductive Reasoning with Language Modelscited by 120
  35. 2023Learning Performance-Improving Code Editscited by 106
  36. 2023Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Modelscited by 67
  37. 2023CodeChain: Towards Modular Code Generation Through Chain of Self-revisions with Representative Sub-modulescited by 58
  38. 2023Prompt Engineering a Prompt Engineercited by 53

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 22; 2025: 73; 2026: 178.

273 papers.

Sort:

By area:

  1. 2024AFlow: Automating Agentic Workflow Generationcited by 432
  2. 2024Automated Design of Agentic Systemscited by 407
  3. 2025Agentic Context Engineering: Evolving Contexts for Self-Improving Language Modelscited by 398
  4. 2025Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agentscited by 356
  5. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligencecited by 170
  6. 2026SkillOpt: Executive Strategy for Self-Evolving Agent Skillscited by 148
  7. 2025Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systemscited by 126
  8. 2024Trace is the Next AutoDiff: Generative Optimization with Rich Feedback, Execution Traces, and LLMscited by 69
  9. 2025WoW: Towards a World omniscient World model Through Embodied Interactioncited by 67
  10. 2025FlowReasoner: Reinforcing Query-Level Meta-Agentscited by 63
  11. 2025ScoreFlow: Mastering LLM Agent Workflows via Score-based Preference Optimizationcited by 56
  12. 2026Expanding the Capabilities of Reinforcement Learning via Text Feedbackcited by 53
38 more
  1. 2025FLEX: Continuous Agent Evolution via Forward Learning from Experiencecited by 52
  2. 2024Critic-V: VLM Critics Help Catch VLM Errors in Multimodal Reasoningcited by 50
  3. 2026A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agentscited by 50
  4. 2026HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundrycited by 47
  5. 2025SiriuS: Self-improving Multi-agent Systems via Bootstrapped Reasoningcited by 44
  6. 2025Feedback Descent: Open-Ended Text Optimization via Pairwise Comparisoncited by 44
  7. 2024Benchmarking Agentic Workflow Generationcited by 37
  8. 2025EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systemscited by 37
  9. 2025Optimizing Model Selection for Compound AI Systemscited by 36
  10. 2026Recursive Harness Self-Improvementcited by 32
  11. 2025Adversarial Reasoning at Jailbreaking Timecited by 29
  12. 2025AutoPDL: Automatic Prompt Optimization for LLM Agentscited by 27
  13. 2024Aviary: training language agents on challenging scientific taskscited by 26
  14. 2024VERA: Explainable Video Anomaly Detection via Verbalized Learning of Vision-Language Modelscited by 24
  15. 2026ReCreate: Reasoning and Creating Domain Agents Driven by Experiencecited by 24
  16. 2025Weak-for-Strong: Training Weak Meta-Agent to Harness Strong Executorscited by 22
  17. 2026Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verificationcited by 22
  18. 2024REVOLVE: Optimizing AI Systems by Tracking Response Evolution in Textual Optimizationcited by 20
  19. 2025Minions: Cost-efficient Collaboration Between On-device and Cloud Language Modelscited by 20
  20. 2025RobustFlow: Towards Robust Agentic Workflow Generationcited by 20
  21. 2025Maestro: Self-Improving Text-to-Image Generation via Agent Orchestrationcited by 19
  22. 2026Self-Improvements in Modern Agentic Systems: A Surveycited by 19
  23. 2024How to Correctly do Semantic Backpropagation on Language-based Agentic Systemscited by 18
  24. 2025On the Effect of Sampling Diversity in Scaling LLM Inferencecited by 18
  25. 2025Text2Grad: Reinforcement Learning from Natural Language Feedbackcited by 18
  26. 2025Toward Effective Tool-Integrated Reasoning via Self-Evolved Preference Learningcited by 17
  27. 2025Can Textual Gradient Work in Federated Learning?cited by 15
  28. 2026AutoSaddler: Automatic Harness Optimization with Durable Updates from Agent Execution Tracescited by 15
  29. 2026InT: Self-Proposed Interventions Enable Credit Assignment in LLM Reasoningcited by 14
  30. 2025Agentic Learner with Grow-and-Refine Multimodal Semantic Memorycited by 13
  31. 2026PACE: Anytime-Valid Acceptance Tests for Self-Evolving Agentscited by 13
  32. 2026DarwinX: Evolving Agent Harnesses Through Natural Selectioncited by 13
  33. 2024GReaTer: Gradients over Reasoning Makes Smaller Language Models Strong Prompt Optimizerscited by 12
  34. 2025Generating Symbolic World Models via Test-time Scaling of Large Language Modelscited by 12
  35. 2026KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learningcited by 12
  36. 2026VQQA: An Agentic Approach for Video Evaluation and Quality Improvementcited by 12
  37. 2026FALAT: Tracing Failures in LLM Agent Trajectories via Dependency-Guided Searchcited by 12
  38. 2024Improving Parallel Program Performance with LLM Optimizers via Agent-System Interfacescited by 11

50 most cited of 273 All 273 (JSON)