pat-jj/Awesome-Adaptation-of-Agentic-AI

Repo for "Adaptation of Agentic AI"

683

62 commits

updated Jun 21, 2026

See the code

README

Awesome Adaptation of Agentic AI

Awesome Stars License: CC BY-NC-ND 4.0 PRWelcome arXiv

A curated list of papers on adaptation strategies of agentic AI systems. This repository accompanies the paper "Adaptation of Agentic AI" (Ongoing Work).

Cite this paper:

@article{jiang2025adaptation,
  title={Adaptation of Agentic AI},
  author={Jiang, Pengcheng and Lin, Jiacheng and Shi, Zhiyi and Wang, Zifeng and He, Luxi and Wu, Yichen and Zhong, Ming and Song, Peiyang and Zhang, Qizheng and Wang, Heng and others},
  journal={arXiv preprint arXiv:2512.16301},
  year={2025}
}

Table of Contents


Agent Adaptation

A1: Tool Execution Signaled Agent Adaptation

Development Timeline:

RL-based Methods

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2025.11OrionarXiv
Paper
IRRetrieversLFM2GRPO
2025.10olmOCR2arXiv
Paper
Code
Document OCRSynthetic Document VerifierQwen2.5-VLSFT, GRPO
2025.10AlphaProofNature’25
Paper
Formal Theorem ProvingLean CompilerTransformer (3B Enc-Dec)SFT, AlphaZero, TTRL
2025.10ToolExpanderarXiv
Paper
Tool-CallingVarious APIsQwen2.5SFT, GRPO
2025.09BFS-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerQwen2.5BFS-Guided AlphaZero-like EI
2025.09WebGen-AgentarXiv
Paper
Code
Website GenerationVLM, GUI Agent, Code ExecutorVarious ModelsSFT, Step-GRPO
2025.09Tool-R1arXiv
Paper
Code
General Tool-Augmented Reasoning, Multimodal QACode Execution, Multimedia ToolsQwen2.5GRPO
2025.08FTRLarXiv
Paper
Code
Multi-Step Tool-UseSimulated APIsQwen3GRPO
2025.08Goedel-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerQwen3SFT, GRPO
2025.07Leanabell-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerQwen2.5SFT, AlphaZero-like EI
2025.06Router-R1NeurIPS'25
Paper
Code
Multi-Round RoutingLLM Routing PoolQwen2.5, LLaMA3.2PPO
2025.05R1-Code-InterpreterarXiv
Paper
Code
CodingCode Execution SandboxQwen2.5GRPO
2025.05Tool-N1arXiv
Paper
Code
Tool-CallingVarious APIsQwen2.5GRPO
2025.04DeepSeek-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerDeepSeek-V2SFT, GRPO
2025.04Kimina-ProverarXiv
Paper
Code
Formal Theorem ProvingLean CompilerLLaMA-2SFT, AlphaZero-like EI
2025.04SQL-R1NeurIPS'25
Paper
Code
Text2SQL SearchSQL EngineQwen2.5, OmniSQLSFT, GRPO
2025.03Rec-R1TMLR'25
Paper
Code
Recommendation OptimizationRecommendation SystemQwen2.5, LLaMA3.2GRPO
2025.03ReZeroarXiv
Paper
Code
Web Search, IRWeb Search EngineLLaMA3.2GRPO
2025.03Code-R1---
Code
CodingCode ExecutorQwen2.5GRPO
2025.02DeepRetrievalCOLM'25
Paper
Code
Web Search, IR, Text2SQLSearch Engine, Retrievers, SQL exec.Qwen2.5, LLaMA3.2PPO, GRPO
2025.01DeepSeek-R1-Zero (Code)Nature
Paper
CodingCode ExecutorDeepSeek-V3-BaseGRPO
2024.10RLEFICML'25
Paper
CodingCode ExecutorLLaMA3.1PPO
2024.08DeepSeek-Prover-V1.5ICLR’25
Paper
Code
Formal Theorem ProvingLean 4 ProverDeepSeek-Prover-V1.5-RLSFT, GRPO
2024.05LeDexNeurIPS'24
Paper
CodingCode ExecutorStarCoder & CodeLlaMASFT, PPO

SFT & DPO Methods

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2024.12AWLICML'25
Paper
Code
Scientific Reasoning,
Adaptive Tool Usage
Scientific SimulatorsLlama-3.1-8B,
Qwen-2.5-{14/32}B
SFT, DPO
2024.10LeReTICLR'25
Paper
Code
IRDense RetrieverLLaMA3, Gemma2DPO-like (IPO)
2024.10ToolFlowNAACL'25
Paper
Tool-CallingVarious APIsLLaMA3.1SFT
2024.06TP-LLaMANeurIPS'24
Paper
Tool-CallingVarious APIsLLaMA2SFT, DPO
2024.05AutoToolsWWW'25
Paper
Code
Automated Tool-CallingVarious APIsGPT4, LLaMA3, MistralSFT
2024.03CYCLEOOPSLA'24
Paper
CodingCode ExecutorCodeGen, StarCoderSFT
2024.02RetPONAACL'25
Paper
Code
IRRetrieverLLaMA2-7BSFT, DPO
2024.02CodeActICML'24
Paper
Code
CodingCode ExecutorLLaMA2, MistralSFT
2024.01NExTICML'24
Paper
Program RepairCode ExecutorPaLM2SFT
2023.07ToolLLMICLR'24
Paper
Code
Tool-Calling, API Planning, Multi-Tool ReasoningReal-World APIsLLaMA, VicunaSFT
2023.06ToolAlpacaarXiv
Paper
Code
Multi-Turn Tool-UseSimulated APIsVicunaSFT
2023.05GorillaNeurIPS'24
Paper
Code
Tool-Calling, API RetrievalVarious APIsLLaMASFT
2023.05TRICENAACL'24
Paper
Code
Math Reasoning, QA, Multilingual QA, Knowledge RetrievalCalculator, WikiSearch, Atlas QA Model, NLLB TranslatorChatGLM, Alpaca, VicunaSFT
2023.02ToolformerNeurIPS'23
Paper
Code
QA, MathCalculator, QA system, Search Engine, Translation System, CalendarGPT-JSFT



A2: Agent Output Signaled Agent Adaptation

Development Timeline:

Methods with Tools

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2025.10TT-SIarXiv
Paper
Tool CallingVarious APIsQwen2.5Test-Time Fine-Tuning
2025.10A²FMarXiv
Paper
Code
Web Navigation, Math, QASearch Engine, Crawl, Code ExecutorQwen2.5APO, GRPO
2025.09VerlToolarXiv
Paper
Code
Math, QA, SQL, Visual, Web Search, CodingCode Interpreter, Search Engine, SQL Executor, Vision ToolsQwen2.5, Qwen3GRPO
2025.08MedResearcher-R1arXiv
Paper
Code
Medical Multi-hop QAMedical Retriever, Web Search API, Document ReaderMedResearcher-R1SFT, GRPO
2025.08Agent LightningarXiv
Paper
Code
Text-to-SQL, RAG, MathSQL Executor, Retriever, CalculatorLLaMA3.2LightningRL
2025.07CodePRMACL'25
Paper
CodingCode ExecutorQwen2.5-CoderSFT
2025.07DynaSearcherarXiv
Paper
Code
Multi-Hop QA, RAGDocument Search, KG SearchQwen2.5, LLaMA3.1GRPO
2025.06MMSearch-R1arXiv
Paper
Code
Web Browsing, QA, Multimodal SearchImage Search, Web Browsing, RetrieverQwen2.5REINFORCE, SFT
2025.06Self-ChallengingarXiv
Paper
Web Browsing, Calculation, Retail, AirlineCode Interpreter, Web Browser, Database APIsLLaMA3.1REINFORCE, SFT
2025.05StepSearchEMNLP'25
Paper
Code
Multi-Hop QASearch Engine, RetrieverQwen2.5StePPO
2025.05ZeroSearcharXiv
Paper
Code
Multi-Hop QA, QASearch Engine, Web SearchQwen2.5, LLaMA3.2REINFORCE, GPRO, PPO, SFT
2025.05AutoRefineNeurIPS'25
Paper
Code
Multi-Hop QA, QARetrieverQwen2.5GRPO
2025.04ReToolarXiv
Paper
Code
MathCode InterpreterQwen2.5PPO
2025.04ToolRLarXiv
Paper
Code
Tool CallingVarious ToolsVarious ModelsGRPO
2025.04DeepResearcherarXiv
Paper
Code
QA, Multi-Hop Reasoning, Deep ResearchWeb Search API, Web BrowserQwen2.5GRPO
2025.03ReSearchNeurIPS'25
Paper
Code
QASearch Engine, RetrieverQwen2.5GRPO
2025.03Search-R1COLM'25
Paper
Code
QASearch Engine, RetrieverQwen2.5PPO, GRPO
2025.03R1-SearcherarXiv
Paper
Code
QARetrieverLLaMA3.1, Qwen2.5REINFORCE++
2025.02RASarXiv
Paper
Code
QARetrieverLLaMA2, LLaMA3.2SFT
2025.01Agent-RarXiv
Paper
Code
Various TasksMonte Carlo Tree SearchQwen2.5, LLaMA3.2SFT
2024.06Re-ReSTEMNLP'24
Paper
Code
Multi-Hop QA, VQA, Sequential Decision, CodingVarious APIsVarious ModelsDPO
2024.06RPGEMNLP'24
Paper
Code
RAG, QA, Multi-hop ReasoningSearch Engine, RetrieverLLaMA2, GPT3.5SFT
2023.10Self-RAGICLR'24
Paper
Code
RAG, QA, Fact VerificationRetrieverLLaMA2SFT
2023.10FireActarXiv
Paper
Code
QASearch APIGPT3.5, LLaMA2, CodeLLaMASFT

Methods without Tools

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2025.10EmpowerarXiv
Paper
Code
Coding---Gemma3SFT
2025.10KnowRLarXiv
Paper
Code
Knowledge calibration---LLaMA3.1, Qwen2.5REINFORCE++
2025.10GRACEarXiv
Paper
Code
Embedding Tasks---Qwen2.5, Qwen3, LLaMA3.2GRPO
2025.09PEvoEMNLP'25
Paper
Crowd Simulation, Behavioral Alignment---Various LLMsPersona Refinement
2025.06MagistralarXiv
Paper
Math, Coding---MagistralPPO, GRPO
2025.05EHRMindarXiv
Paper
Code
EHR-based Reasoning---LLaMA3SFT, GRPO
2025.01Kimi k1.5arXiv
Paper
Code
Math, Coding---Kimi k1.5GRPO
2025.01DeepSeek-R1-Zero (Math)Nature
Paper
Math---DeepSeek-V3GRPO
2024.09SCoReICLR'25
Paper
Code
Math, Coding, QA---Gemini1.0 Pro, Gemini1.5 FlashREINFORCE
2024.07RISENeurIPS'24
Paper
Code
Math---LLaMA2, LLaMA3, MistralSFT
2024.06TextGradNature
Paper
Code
Various Tasks---GPT3.5, GPT4oPrompt Tuning
2023.03Self-RefineNeurIPS'23
Paper
Code
Dialogue, Math, Coding---GPT3.5, GPT4, CODEXTest-Time Prompting

Tool Adaptation

T1: Agent-Agnostic Tool Adaptation

Foundational Systems and Architectures

Year.MonthMethod NameVenuePaper Name
2021.08Neural OperatorsJMLR'23
Paper
Neural Operator: Learning Maps Between Function Spaces
2023.09HuggingGPTNeurIPS'23
Paper
Code
HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
2023.08ViperGPTICCV'23
Paper
Code
ViperGPT: Visual Inference via Python Execution for Reasoning
2025.07SciToolAgentNature Comp. Sci.'25
Paper
SciToolAgent: A Knowledge-Graph-Driven Scientific Agent for Multitool Integration

Categories and Training Methods

Year.MonthMethod NameVenuePaper Name
2021.01CLIPICML'21
Paper
Code
Learning Transferable Visual Models from Natural Language Supervision
2023.04SAMICCV'23
Paper
Code
Segment Anything
2024.06SAM-CLIPCVPR'24
Paper
SAM-CLIP: Merging Vision Foundation Models Towards Semantic and Spatial Understanding
2023.12WhisperICML'23
Paper
Code
Robust Speech Recognition via Large-Scale Weak Supervision
2024.10CodeActICML'24
Paper
Code
Executable Code Actions Elicit Better LLM Agents
2020.04DPREMNLP'20
Paper
Code
Dense Passage Retrieval for Open-Domain Question Answering
2020.04ColBERTSIGIR'20
Paper
Code
ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT
2021.12ContrieverTMLR'22
Paper
Code
Unsupervised Dense Information Retrieval with Contrastive Learning
2022.12e5arXiv
Paper
Code
Text Embeddings by Weakly-Supervised Contrastive Pre-training
2021.07AlphaFold2Nature
Paper
Code
Highly Accurate Protein Structure Prediction with AlphaFold
2023.03ESMFoldScience
Paper
Evolutionary-Scale Prediction of Atomic-Level Protein Structure with a Language Model



T2: Agent-Supervised Tool Adaptation

Development Timeline:

TimeMethodVenueTask(s)Tool BackboneAgent BackboneTuning
2025.10QAgentarXiv
Paper
Code
QA, RAGQwen2.5-3BQwen-7BGRPO
2025.10AgentFlowarXiv
Paper
Code
Web Search, Planning, Reasoning, MathQwen2.5-7BQwen2.5-7BFlow-GRPO
2025.10Advisor ModelsarXiv
Paper
Code
Math, ReasoningQwen2.5-7B, Qwen3-8BGPT-4o-Mini, GPT-5, Claude4-Sonnet, GPT-4.1-MiniGRPO
2025.10AutoGraph-R1arXiv
Paper
Code
KG Construction, RAGKG Constructor (Qwen2.5-3B/7B)Frozen RAG Generator (Qwen2.5-7B)GRPO
2025.10MAEarXiv
Paper
Code
Math, Coding, Commonsense ReasoningQwen2.5-3BQwen2.5-3BREINFORCE++
2025.09Mem-αarXiv
Paper
Code
Retrieval, Test-Time Learning, Long-Range UnderstandingQwen3-4BQwen3-4B, Qwen3-32B, GPT-4.1-MiniGRPO
2025.08AI-SearchPlannerarXiv
Paper
Web QAQwen3-32bQwen2.5-7BPPO
2025.08MementoarXiv
Paper
Code
Long-Horizon Reasoning, Web Research, QA, Academic ReasoningQ-function (two-layer MLPs)GPT-4.1Soft Q-Learning
2025.08R-ZeroarXiv
Paper
Code
Math, ReasoningQwen3-4B, Qwen3-8B, OctoThinker-3B, OctoThinker-8BQwen3-4B, Qwen3-8B, OctoThinker-3B, OctoThinker-8BGRPO
2025.06SysformerarXiv
Paper
QA, RAGSmall TransformerLLaMA-2-7B, LLaMA-3.1-8B, Mistral-7B, Phi-3.5-mini, Zephyr-7B-betaSupervised Learning
2025.05s3EMNLP'25
Paper
Code
QA, RAGQwen2.5-7BQwen2.5-7B, Qwen2.5-14B, Claude-3-HaikuPPO
2024.10Matryoshka PilotNeurIPS'25
Paper
Code
Math, Planning, ReasoningLLaMA3-8B, Qwen2.5-7BGPT-4o-Mini, GPT-3.5-TurboDPO, IDPO
2024.06CoBBEMNLP'24
Paper
Code
QA, MathMistral-7b-inst-v2GPT-3.5-Turbo, Claude-3-Haiku, Phi-3-mini-4k-inst, Gemma-1.1-7B-it, Mistral-7B-inst-v2SFT, ORPO
2024.05MedadapterEMNLP'24
Paper
Code
Medical QA, NLI, RQEBERT-Base-UncasedGPT-3.5-TurboSFT, BPO
2024.03BLADEAAAI'25
Paper
Code
Domain-Specific QABLOOMZ-1b7ChatGPT, ChatGLM, Baichuan, QwenSFT, BPO
2024.02ARL2ACL'24
Paper
Code
QALLaMA2-7BGPT-3.5-TurboContrastive Learning
2024.02EVOREMNLP'24
Paper
Code
RAG-based CodingGPT-3.5-TurboGPT-3.5-Turbo, CodeLLaMAPrompt Engineering
2024.02Bbox-AdapterICML'24
Paper
Code
QADeBERTa-v3-base (0.1B), DeBERTa-v3-large (0.3B)GPT-3.5-Turbo, Mixtral-8x7BContrastive Learning
2024.01Proxy-TuningCOLM'24
Paper
Code
QA, Math, CodeLLaMA2-7BLLaMA2-70BProxy-Tuning
2024.01BGMACL'24
Paper
QA, Personalized Generation (NQ, HotpotQA, Email, Book)T5-XXL-11BPaLM2-SSFT, PPO
2023.10RA-DITICLR'24
Paper
Knowledge-Intensive Tasks (MMLU, NQ, TQA, ELI5, HotpotQA, etc.)DRAGON+LLaMA-65BSFT, LSR
2023.06LLM-REACL'24
Paper
Code
Zero-shot NLU (Reading Comprehension, QA, NLI, Paraphrase, Sentiment, Summarization)E5-baseGPT-Neo-2.7B, LLaMA-13B, GPT-3.5-TurboContrastive Learning
2023.05AARACL'23
Paper
Code
Zero-Shot Generalization (MMLU, PopQA)ANCE, ContrieverFlan-T5-Small, InstructGPTContrastive Learning
2023.05ToolkenGPTNeurIPS'23
Paper
Code
Numerical Reasoning, QA, Plan GenerationToken EmbeddingGPT-J 6B, OPT-6.7B, OPT-13BProxy-Tuning
2023.03UPRISEEMNLP'23
Paper
Code
Zero-shot NLU (Reading Comprehension, QA, NLI, Paraphrase, Sentiment, Summarization)GPT-Neo-2.7BBLOOM-7.1B, OPT-66B, GPT-3-175BContrastive Learning
2023.01REPLUGNAACL'24
Paper
Code
QAContrieverGPT3-175B, PaLM, Codex, LLaMA-13BProxy-Tuning, LSR

Citation

If you find this repository useful, please consider citing our survey:

@article{jiang2025adaptation,
  title={Adaptation of Agentic AI},
  author={Jiang, Pengcheng and Lin, Jiacheng and Shi, Zhiyi and Wang, Zifeng and He, Luxi and Wu, Yichen and Zhong, Ming and Song, Peiyang and Zhang, Qizheng and Wang, Heng and others},
  journal={arXiv preprint arXiv:2512.16301},
  year={2025}
}

Contributing

We welcome contributions! Please feel free to submit a Pull Request to add new papers or update existing entries.



(ノ◕ヮ◕)ノ*:・゚✧ Keep exploring the awesome world of agentic AI! ✧゚・: *ヽ(◕ヮ◕ヽ)

adaptation
agentic-ai
large-language-models

Contributors

pat-jj

47 commits

zhiyiscs

7 commits

linjc16

4 commits

Peiyang-Song

2 commits

pat-jj/Awesome-Adaptation-of-Agentic-AI

Repo for "Adaptation of Agentic AI"

683

62 commits

updated Jun 21, 2026

See the code

README

Awesome Adaptation of Agentic AI

Awesome Stars License: CC BY-NC-ND 4.0 PRWelcome arXiv

A curated list of papers on adaptation strategies of agentic AI systems. This repository accompanies the paper "Adaptation of Agentic AI" (Ongoing Work).

Cite this paper:

@article{jiang2025adaptation,
  title={Adaptation of Agentic AI},
  author={Jiang, Pengcheng and Lin, Jiacheng and Shi, Zhiyi and Wang, Zifeng and He, Luxi and Wu, Yichen and Zhong, Ming and Song, Peiyang and Zhang, Qizheng and Wang, Heng and others},
  journal={arXiv preprint arXiv:2512.16301},
  year={2025}
}

Table of Contents


Agent Adaptation

A1: Tool Execution Signaled Agent Adaptation

Development Timeline:

RL-based Methods

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2025.11OrionarXiv
Paper
IRRetrieversLFM2GRPO
2025.10olmOCR2arXiv
Paper
Code
Document OCRSynthetic Document VerifierQwen2.5-VLSFT, GRPO
2025.10AlphaProofNature’25
Paper
Formal Theorem ProvingLean CompilerTransformer (3B Enc-Dec)SFT, AlphaZero, TTRL
2025.10ToolExpanderarXiv
Paper
Tool-CallingVarious APIsQwen2.5SFT, GRPO
2025.09BFS-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerQwen2.5BFS-Guided AlphaZero-like EI
2025.09WebGen-AgentarXiv
Paper
Code
Website GenerationVLM, GUI Agent, Code ExecutorVarious ModelsSFT, Step-GRPO
2025.09Tool-R1arXiv
Paper
Code
General Tool-Augmented Reasoning, Multimodal QACode Execution, Multimedia ToolsQwen2.5GRPO
2025.08FTRLarXiv
Paper
Code
Multi-Step Tool-UseSimulated APIsQwen3GRPO
2025.08Goedel-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerQwen3SFT, GRPO
2025.07Leanabell-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerQwen2.5SFT, AlphaZero-like EI
2025.06Router-R1NeurIPS'25
Paper
Code
Multi-Round RoutingLLM Routing PoolQwen2.5, LLaMA3.2PPO
2025.05R1-Code-InterpreterarXiv
Paper
Code
CodingCode Execution SandboxQwen2.5GRPO
2025.05Tool-N1arXiv
Paper
Code
Tool-CallingVarious APIsQwen2.5GRPO
2025.04DeepSeek-Prover-V2arXiv
Paper
Code
Formal Theorem ProvingLean CompilerDeepSeek-V2SFT, GRPO
2025.04Kimina-ProverarXiv
Paper
Code
Formal Theorem ProvingLean CompilerLLaMA-2SFT, AlphaZero-like EI
2025.04SQL-R1NeurIPS'25
Paper
Code
Text2SQL SearchSQL EngineQwen2.5, OmniSQLSFT, GRPO
2025.03Rec-R1TMLR'25
Paper
Code
Recommendation OptimizationRecommendation SystemQwen2.5, LLaMA3.2GRPO
2025.03ReZeroarXiv
Paper
Code
Web Search, IRWeb Search EngineLLaMA3.2GRPO
2025.03Code-R1---
Code
CodingCode ExecutorQwen2.5GRPO
2025.02DeepRetrievalCOLM'25
Paper
Code
Web Search, IR, Text2SQLSearch Engine, Retrievers, SQL exec.Qwen2.5, LLaMA3.2PPO, GRPO
2025.01DeepSeek-R1-Zero (Code)Nature
Paper
CodingCode ExecutorDeepSeek-V3-BaseGRPO
2024.10RLEFICML'25
Paper
CodingCode ExecutorLLaMA3.1PPO
2024.08DeepSeek-Prover-V1.5ICLR’25
Paper
Code
Formal Theorem ProvingLean 4 ProverDeepSeek-Prover-V1.5-RLSFT, GRPO
2024.05LeDexNeurIPS'24
Paper
CodingCode ExecutorStarCoder & CodeLlaMASFT, PPO

SFT & DPO Methods

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2024.12AWLICML'25
Paper
Code
Scientific Reasoning,
Adaptive Tool Usage
Scientific SimulatorsLlama-3.1-8B,
Qwen-2.5-{14/32}B
SFT, DPO
2024.10LeReTICLR'25
Paper
Code
IRDense RetrieverLLaMA3, Gemma2DPO-like (IPO)
2024.10ToolFlowNAACL'25
Paper
Tool-CallingVarious APIsLLaMA3.1SFT
2024.06TP-LLaMANeurIPS'24
Paper
Tool-CallingVarious APIsLLaMA2SFT, DPO
2024.05AutoToolsWWW'25
Paper
Code
Automated Tool-CallingVarious APIsGPT4, LLaMA3, MistralSFT
2024.03CYCLEOOPSLA'24
Paper
CodingCode ExecutorCodeGen, StarCoderSFT
2024.02RetPONAACL'25
Paper
Code
IRRetrieverLLaMA2-7BSFT, DPO
2024.02CodeActICML'24
Paper
Code
CodingCode ExecutorLLaMA2, MistralSFT
2024.01NExTICML'24
Paper
Program RepairCode ExecutorPaLM2SFT
2023.07ToolLLMICLR'24
Paper
Code
Tool-Calling, API Planning, Multi-Tool ReasoningReal-World APIsLLaMA, VicunaSFT
2023.06ToolAlpacaarXiv
Paper
Code
Multi-Turn Tool-UseSimulated APIsVicunaSFT
2023.05GorillaNeurIPS'24
Paper
Code
Tool-Calling, API RetrievalVarious APIsLLaMASFT
2023.05TRICENAACL'24
Paper
Code
Math Reasoning, QA, Multilingual QA, Knowledge RetrievalCalculator, WikiSearch, Atlas QA Model, NLLB TranslatorChatGLM, Alpaca, VicunaSFT
2023.02ToolformerNeurIPS'23
Paper
Code
QA, MathCalculator, QA system, Search Engine, Translation System, CalendarGPT-JSFT



A2: Agent Output Signaled Agent Adaptation

Development Timeline:

Methods with Tools

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2025.10TT-SIarXiv
Paper
Tool CallingVarious APIsQwen2.5Test-Time Fine-Tuning
2025.10A²FMarXiv
Paper
Code
Web Navigation, Math, QASearch Engine, Crawl, Code ExecutorQwen2.5APO, GRPO
2025.09VerlToolarXiv
Paper
Code
Math, QA, SQL, Visual, Web Search, CodingCode Interpreter, Search Engine, SQL Executor, Vision ToolsQwen2.5, Qwen3GRPO
2025.08MedResearcher-R1arXiv
Paper
Code
Medical Multi-hop QAMedical Retriever, Web Search API, Document ReaderMedResearcher-R1SFT, GRPO
2025.08Agent LightningarXiv
Paper
Code
Text-to-SQL, RAG, MathSQL Executor, Retriever, CalculatorLLaMA3.2LightningRL
2025.07CodePRMACL'25
Paper
CodingCode ExecutorQwen2.5-CoderSFT
2025.07DynaSearcherarXiv
Paper
Code
Multi-Hop QA, RAGDocument Search, KG SearchQwen2.5, LLaMA3.1GRPO
2025.06MMSearch-R1arXiv
Paper
Code
Web Browsing, QA, Multimodal SearchImage Search, Web Browsing, RetrieverQwen2.5REINFORCE, SFT
2025.06Self-ChallengingarXiv
Paper
Web Browsing, Calculation, Retail, AirlineCode Interpreter, Web Browser, Database APIsLLaMA3.1REINFORCE, SFT
2025.05StepSearchEMNLP'25
Paper
Code
Multi-Hop QASearch Engine, RetrieverQwen2.5StePPO
2025.05ZeroSearcharXiv
Paper
Code
Multi-Hop QA, QASearch Engine, Web SearchQwen2.5, LLaMA3.2REINFORCE, GPRO, PPO, SFT
2025.05AutoRefineNeurIPS'25
Paper
Code
Multi-Hop QA, QARetrieverQwen2.5GRPO
2025.04ReToolarXiv
Paper
Code
MathCode InterpreterQwen2.5PPO
2025.04ToolRLarXiv
Paper
Code
Tool CallingVarious ToolsVarious ModelsGRPO
2025.04DeepResearcherarXiv
Paper
Code
QA, Multi-Hop Reasoning, Deep ResearchWeb Search API, Web BrowserQwen2.5GRPO
2025.03ReSearchNeurIPS'25
Paper
Code
QASearch Engine, RetrieverQwen2.5GRPO
2025.03Search-R1COLM'25
Paper
Code
QASearch Engine, RetrieverQwen2.5PPO, GRPO
2025.03R1-SearcherarXiv
Paper
Code
QARetrieverLLaMA3.1, Qwen2.5REINFORCE++
2025.02RASarXiv
Paper
Code
QARetrieverLLaMA2, LLaMA3.2SFT
2025.01Agent-RarXiv
Paper
Code
Various TasksMonte Carlo Tree SearchQwen2.5, LLaMA3.2SFT
2024.06Re-ReSTEMNLP'24
Paper
Code
Multi-Hop QA, VQA, Sequential Decision, CodingVarious APIsVarious ModelsDPO
2024.06RPGEMNLP'24
Paper
Code
RAG, QA, Multi-hop ReasoningSearch Engine, RetrieverLLaMA2, GPT3.5SFT
2023.10Self-RAGICLR'24
Paper
Code
RAG, QA, Fact VerificationRetrieverLLaMA2SFT
2023.10FireActarXiv
Paper
Code
QASearch APIGPT3.5, LLaMA2, CodeLLaMASFT

Methods without Tools

TimeMethodVenueTask(s)Tool(s)Agent BackboneTuning
2025.10EmpowerarXiv
Paper
Code
Coding---Gemma3SFT
2025.10KnowRLarXiv
Paper
Code
Knowledge calibration---LLaMA3.1, Qwen2.5REINFORCE++
2025.10GRACEarXiv
Paper
Code
Embedding Tasks---Qwen2.5, Qwen3, LLaMA3.2GRPO
2025.09PEvoEMNLP'25
Paper
Crowd Simulation, Behavioral Alignment---Various LLMsPersona Refinement
2025.06MagistralarXiv
Paper
Math, Coding---MagistralPPO, GRPO
2025.05EHRMindarXiv
Paper
Code
EHR-based Reasoning---LLaMA3SFT, GRPO
2025.01Kimi k1.5arXiv
Paper
Code
Math, Coding---Kimi k1.5GRPO
2025.01DeepSeek-R1-Zero (Math)Nature
Paper
Math---DeepSeek-V3GRPO
2024.09SCoReICLR'25
Paper
Code
Math, Coding, QA---Gemini1.0 Pro, Gemini1.5 FlashREINFORCE
2024.07RISENeurIPS'24
Paper
Code
Math---LLaMA2, LLaMA3, MistralSFT
2024.06TextGradNature
Paper
Code
Various Tasks---GPT3.5, GPT4oPrompt Tuning
2023.03Self-RefineNeurIPS'23
Paper
Code
Dialogue, Math, Coding---GPT3.5, GPT4, CODEXTest-Time Prompting

Tool Adaptation

T1: Agent-Agnostic Tool Adaptation

Foundational Systems and Architectures

Year.MonthMethod NameVenuePaper Name
2021.08Neural OperatorsJMLR'23
Paper
Neural Operator: Learning Maps Between Function Spaces
2023.09HuggingGPTNeurIPS'23
Paper
Code
HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
2023.08ViperGPTICCV'23
Paper
Code
ViperGPT: Visual Inference via Python Execution for Reasoning
2025.07SciToolAgentNature Comp. Sci.'25
Paper
SciToolAgent: A Knowledge-Graph-Driven Scientific Agent for Multitool Integration

Categories and Training Methods

Year.MonthMethod NameVenuePaper Name
2021.01CLIPICML'21
Paper
Code
Learning Transferable Visual Models from Natural Language Supervision
2023.04SAMICCV'23
Paper
Code
Segment Anything
2024.06SAM-CLIPCVPR'24
Paper
SAM-CLIP: Merging Vision Foundation Models Towards Semantic and Spatial Understanding
2023.12WhisperICML'23
Paper
Code
Robust Speech Recognition via Large-Scale Weak Supervision
2024.10CodeActICML'24
Paper
Code
Executable Code Actions Elicit Better LLM Agents
2020.04DPREMNLP'20
Paper
Code
Dense Passage Retrieval for Open-Domain Question Answering
2020.04ColBERTSIGIR'20
Paper
Code
ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT
2021.12ContrieverTMLR'22
Paper
Code
Unsupervised Dense Information Retrieval with Contrastive Learning
2022.12e5arXiv
Paper
Code
Text Embeddings by Weakly-Supervised Contrastive Pre-training
2021.07AlphaFold2Nature
Paper
Code
Highly Accurate Protein Structure Prediction with AlphaFold
2023.03ESMFoldScience
Paper
Evolutionary-Scale Prediction of Atomic-Level Protein Structure with a Language Model



T2: Agent-Supervised Tool Adaptation

Development Timeline:

TimeMethodVenueTask(s)Tool BackboneAgent BackboneTuning
2025.10QAgentarXiv
Paper
Code
QA, RAGQwen2.5-3BQwen-7BGRPO
2025.10AgentFlowarXiv
Paper
Code
Web Search, Planning, Reasoning, MathQwen2.5-7BQwen2.5-7BFlow-GRPO
2025.10Advisor ModelsarXiv
Paper
Code
Math, ReasoningQwen2.5-7B, Qwen3-8BGPT-4o-Mini, GPT-5, Claude4-Sonnet, GPT-4.1-MiniGRPO
2025.10AutoGraph-R1arXiv
Paper
Code
KG Construction, RAGKG Constructor (Qwen2.5-3B/7B)Frozen RAG Generator (Qwen2.5-7B)GRPO
2025.10MAEarXiv
Paper
Code
Math, Coding, Commonsense ReasoningQwen2.5-3BQwen2.5-3BREINFORCE++
2025.09Mem-αarXiv
Paper
Code
Retrieval, Test-Time Learning, Long-Range UnderstandingQwen3-4BQwen3-4B, Qwen3-32B, GPT-4.1-MiniGRPO
2025.08AI-SearchPlannerarXiv
Paper
Web QAQwen3-32bQwen2.5-7BPPO
2025.08MementoarXiv
Paper
Code
Long-Horizon Reasoning, Web Research, QA, Academic ReasoningQ-function (two-layer MLPs)GPT-4.1Soft Q-Learning
2025.08R-ZeroarXiv
Paper
Code
Math, ReasoningQwen3-4B, Qwen3-8B, OctoThinker-3B, OctoThinker-8BQwen3-4B, Qwen3-8B, OctoThinker-3B, OctoThinker-8BGRPO
2025.06SysformerarXiv
Paper
QA, RAGSmall TransformerLLaMA-2-7B, LLaMA-3.1-8B, Mistral-7B, Phi-3.5-mini, Zephyr-7B-betaSupervised Learning
2025.05s3EMNLP'25
Paper
Code
QA, RAGQwen2.5-7BQwen2.5-7B, Qwen2.5-14B, Claude-3-HaikuPPO
2024.10Matryoshka PilotNeurIPS'25
Paper
Code
Math, Planning, ReasoningLLaMA3-8B, Qwen2.5-7BGPT-4o-Mini, GPT-3.5-TurboDPO, IDPO
2024.06CoBBEMNLP'24
Paper
Code
QA, MathMistral-7b-inst-v2GPT-3.5-Turbo, Claude-3-Haiku, Phi-3-mini-4k-inst, Gemma-1.1-7B-it, Mistral-7B-inst-v2SFT, ORPO
2024.05MedadapterEMNLP'24
Paper
Code
Medical QA, NLI, RQEBERT-Base-UncasedGPT-3.5-TurboSFT, BPO
2024.03BLADEAAAI'25
Paper
Code
Domain-Specific QABLOOMZ-1b7ChatGPT, ChatGLM, Baichuan, QwenSFT, BPO
2024.02ARL2ACL'24
Paper
Code
QALLaMA2-7BGPT-3.5-TurboContrastive Learning
2024.02EVOREMNLP'24
Paper
Code
RAG-based CodingGPT-3.5-TurboGPT-3.5-Turbo, CodeLLaMAPrompt Engineering
2024.02Bbox-AdapterICML'24
Paper
Code
QADeBERTa-v3-base (0.1B), DeBERTa-v3-large (0.3B)GPT-3.5-Turbo, Mixtral-8x7BContrastive Learning
2024.01Proxy-TuningCOLM'24
Paper
Code
QA, Math, CodeLLaMA2-7BLLaMA2-70BProxy-Tuning
2024.01BGMACL'24
Paper
QA, Personalized Generation (NQ, HotpotQA, Email, Book)T5-XXL-11BPaLM2-SSFT, PPO
2023.10RA-DITICLR'24
Paper
Knowledge-Intensive Tasks (MMLU, NQ, TQA, ELI5, HotpotQA, etc.)DRAGON+LLaMA-65BSFT, LSR
2023.06LLM-REACL'24
Paper
Code
Zero-shot NLU (Reading Comprehension, QA, NLI, Paraphrase, Sentiment, Summarization)E5-baseGPT-Neo-2.7B, LLaMA-13B, GPT-3.5-TurboContrastive Learning
2023.05AARACL'23
Paper
Code
Zero-Shot Generalization (MMLU, PopQA)ANCE, ContrieverFlan-T5-Small, InstructGPTContrastive Learning
2023.05ToolkenGPTNeurIPS'23
Paper
Code
Numerical Reasoning, QA, Plan GenerationToken EmbeddingGPT-J 6B, OPT-6.7B, OPT-13BProxy-Tuning
2023.03UPRISEEMNLP'23
Paper
Code
Zero-shot NLU (Reading Comprehension, QA, NLI, Paraphrase, Sentiment, Summarization)GPT-Neo-2.7BBLOOM-7.1B, OPT-66B, GPT-3-175BContrastive Learning
2023.01REPLUGNAACL'24
Paper
Code
QAContrieverGPT3-175B, PaLM, Codex, LLaMA-13BProxy-Tuning, LSR

Citation

If you find this repository useful, please consider citing our survey:

@article{jiang2025adaptation,
  title={Adaptation of Agentic AI},
  author={Jiang, Pengcheng and Lin, Jiacheng and Shi, Zhiyi and Wang, Zifeng and He, Luxi and Wu, Yichen and Zhong, Ming and Song, Peiyang and Zhang, Qizheng and Wang, Heng and others},
  journal={arXiv preprint arXiv:2512.16301},
  year={2025}
}

Contributing

We welcome contributions! Please feel free to submit a Pull Request to add new papers or update existing entries.



(ノ◕ヮ◕)ノ*:・゚✧ Keep exploring the awesome world of agentic AI! ✧゚・: *ヽ(◕ヮ◕ヽ)

adaptation
agentic-ai
large-language-models

Contributors

pat-jj

47 commits

zhiyiscs

7 commits

linjc16

4 commits

Peiyang-Song

2 commits