AgentFlare

Scholar · tool use and function calling in language models

Tool Use in Language Models: Research Digest

10 foundational papers with 12,411 combined citations, plus the latest arXiv preprints.

Foundational papers1012,411 citations
Most citedToolformer: Language models can teach…7,859 citations · 2023
arXiv, Sep 2026102 papers+96% vs Oct 2025
Newest preprint2026-10-15

AISA Scholar API · arXiv

arXiv submissions per monthPapers matching (abs:"tool use" OR abs:"function calling") AND abs:"language model" · last 12 complete months
monthly arXiv submissions: Oct 2025 52 → Sep 2026 102; high 103 (May 2026), low 41 (Nov 2025)366084108Oct 2025Mar 2026Sep 2026
402 · Pay-per-Crawl$0.02

You're reading the full report — free for humans and search engines. The same content ships as a structured JSON dataset (normalized facts, sources and citation arrays, no HTML parsing) so AI agents can consume it directly — metered per fetch via the open HTTP 402 payment protocol.

⚡ Agents: fetch .data.json · $0.02 → agents: crawlerAutoPrice=true · verify at https://esa.aisa.one/api/v1/access/verify

What’s new on arXiv

A12026-10-15cs.AI

MLExpert: Towards a Generalist Agentic Model for Autonomous Machine Learning Engineering

Shangheng Du, Jinxin Shi, Shuaiyu Zhang, Yiming Wu, Junpeng Zhao, Shiyang Feng, Xin Li, Liang He, Xiangchao Yan, Bo Zhang

Large language models (LLMs) are expanding from assisting scientific research and engineering to autonomously developing and improving AI systems, advancing AI for AI (AI4AI) as a pathway toward recursive self-improvement (RSI). Workflow-driven machine learning engineering (MLE) systems provide a practical path toward this pursuit through complex orchestration that manages long-horizon experimentation, while recent MLE systems adopt model-directed interaction, allowing models to autonomously organize experiments and adapt optimization strategies based on execution feedback. However, feedback from each experiment reflects several preceding decisions, and a long MLE run contains many such costly experiments. Training must therefore select useful supervision from long interaction trajectories and learn from intermediate outcomes. We propose a post-training method for autonomous long-horizon MLE optimization and apply it to train MLExpert, a 35B Mixture-of-Experts agentic model. We design the Kaggle-MLE Factory to construct executable tasks and an interaction harness for model-directed experimentation. Hierarchical agentic SFT transfers general tool-use and coding capabilities to MLE through outcome-guided supervision. Windowed Interaction RL (WIRL) treats the multi-turn decisions between successive experiment results as an execution window and assigns it a shared outcome-based reward. Asynchronous collection and policy correction support policy updates before the trajectory ends. On MLE-Bench Lite, MLExpert improves the average medal rate of the base model from 37.9% to 72.7%, achieving performance comparable to larger frontier models such as Kimi-K3 and GLM-5.3. Further experiments show that MLExpert improves performance on NatureBench Lite, agent scaffold optimization, and algorithm optimization, suggesting that the learned optimization capability generalizes to other tasks.

A22026-10-15cs.LG

Learning to Steer LLM Agents and Multi-Agent Systems

Leyang Shen, Yaxin Gong, Yang Zhang, Chenxiao Fan, Jiakai Wang, Dong Wang, Yang Liu, Fuli Feng, Chun Kai Ling, Tat-Seng Chua

Large language model (LLM)-based agents and multi-agent systems increasingly tackle complex tasks through long-horizon tool use and inter-agent collaboration. However, these systems are prone to deviating from the overall goal as execution unfolds, which wastes execution resources and leads to undesired outcomes, a failure termed goal drift. In this work, we propose a steering agent that supervises execution on the fly and steers the system back on track when drift occurs. The steering agent observes the execution context, deciding when to intervene with a low-cost fast path and how to intervene with a high-cost slow path. To unlock its steering intelligence, the ability to assess goal drift and generate effective interventions, we jointly train the two paths with reinforcement learning, combining a pairwise ranking objective with group-relative policy optimization (GRPO). Experiments on information-seeking search tasks show that the steering agent provides effective supervision for both single-agent and multi-agent systems, improving their performance and efficiency. The learned steering intelligence generalizes to out-of-domain tasks, transfers to unseen executors, and supports controllable intervention frequency without retraining.

A32026-10-15cs.AI

PROBE: Learning to Audit Policy Compliance in Tool-Using LLM Agents

Kshitij Mishra, Abhijith Sharma, Nils Lukas, Salem Lahlou

Modern agentic systems extend language models with external tools, enabling them to complete complex tasks through multi-turn interactions. As these agents are deployed in consequential settings, they are often constrained by policies governing pricing, privacy, confirmation, safety, and other requirements. However, existing evaluation methods leave a critical gap: capability benchmarks measure what agents can accomplish, while content-safety red-teaming measures harmful generations, but neither tests whether agents comply with predefined policies during otherwise valid tool use. Such policy violations are difficult to audit because their effects may emerge only after several turns, span multiple tool calls, or surface in downstream systems. We propose PROBE (Policy Rule Observation via Behavioral Exploration), an algorithm for training LLM-based auditors that interact with deployed agents as realistic users and probe them for policy violations through multi-turn, tool-mediated dialogue. We formalize agent auditing as a two-player Markov game and propose a composite reward that balances violation discovery, behavioral plausibility, conversation progress, and policy diversity. To study training objectives and reward design, we compare GRPO, DPO, and GFlowNet-based optimization in a controlled setting. Across extensive experiments, auditors trained with PROBE uncover substantially broader and higher-yield policy violations than same-architecture zero-shot adversaries, even when the auditor is significantly smaller than the target agent. Our results show that multi-turn behavioral auditing is a distinct and necessary axis of agent evaluation, complementing capability and safety benchmarks. PROBE offers a scalable path toward automated policy auditing for real-world agentic systems and suggests a co-evolutionary loop in which auditors expose failures and agents improve from auditor feedback.

A42026-10-15cs.AI

RSIFlow: Adaptive Self-Improvement Workflows with an Evolving Meta-Agent

Huan Zhang, Mingju Chen, Dongxu Zhou, Can Lv, Sen Cui, Faguo Wu, Shiji Zhou

Large language model (LLM) agents have demonstrated strong capabilities in complex tasks, including code generation, tool use, and multi-hop question answering. Existing self-improvement methods enhance agent performance through model optimization, harness evolution, and executable artifact refinement, while recursive self-improvement (RSI) further extends optimization to the improvement mechanisms themselves. However, adapting self-improvement workflows to evolving performance bottlenecks while improving workflow orchestration through execution feedback remains a key challenge. We propose RSIFlow, a Recursive Self-Improvement framework that integrates adaptive workflow orchestration with meta-agent capability evolution. RSIFlow diagnoses performance bottlenecks from task-agent trajectories and environmental feedback to dynamically select component-specific improvement workflows. The meta-agent then distills workflow outcomes into component-specific skills and cross-component principles, evolving its own harness to improve subsequent workflow orchestration. Experiments across six benchmarks and two backbone models show that RSIFlow achieves the SOTA average performance on all benchmarks. Ablation studies further demonstrate the contributions of adaptive workflow orchestration and meta-harness evolution. Code is available at https://github.com/1HUAN1/RSI-Flow.

A52026-10-15cs.CL

Rethinking the Role of Tool Graph Memory in Multi-Turn Tool Use

Uiji Hwang, Richeng Xuan, Taeuk Kim

Large language model (LLM) agents solve complex tasks by calling external tools over multiple turns. Recent work has proposed tool graph memory, which stores past experience as tool transitions to guide subsequent selection. However, it remains unclear whether this component consistently improves agents' tool use or undermines it. We evaluate tool graph memory on nine models and two benchmarks, finding that it seldom improves significantly over baselines, and that its gains, where present, cost over twice the LLM calls. The memory's guidance is often incorrect, and random or no guidance retains most of the gain, implying that the benefit stems from extra computation rather than the memory itself. Moreover, even with the correct next tool suggested, many failures remain, such as giving up without a tool call. Storing the same experience instead as textual insights outperforms tool graph memory with almost no extra LLM calls. This suggests that memory serves better as a soft prior than as a prescription of the next tool, leading to a more effective and generalizable solution.

A62026-10-14cs.CR

P2Patch: Exploit-Guided Vulnerability Repair

Mohsen Iranmanesh, Sina Moradi Sabet, Sina Marefat, Ali Javidi Ghasr, Maxime Newman Nereyabagabo, Kousha Amouzesh, Jaivir Malhi, Vinci Vela…

Large language models are increasingly deployed as autonomous agents for automated program repair, patching software vulnerabilities through multi-step reasoning and tool use. Yet these systems typically validate a patch against the same evidence used to construct it -- a static alert, sanitizer trace, published exploit, or test suite. A patch that blocks that one piece of evidence can therefore look successful while remaining vulnerable to the same vulnerability with some variation. We present P2Patch, a generative adversarial repair system that treats each patch as a new attack target. Starting from a normalized vulnerability finding, an exploitation agent constructs a reproducing proof of vulnerability (PoV), a patching agent repairs the program, and the exploitation agent then attacks the patched program for a distinct same-vulnerability exploit. Every successful exploit becomes a permanent repair requirement, and the final patch must block all PoVs accumulated during the patching process, with construction-time PoVs kept strictly separate from held-out evaluation oracles. Across 101 C/C++ and Java CVE subjects, this adversarial loop surfaces a distinct confirmed exploit in 41% of runs after the patch has already blocked its original construction-time PoV. Against single-pass repair from the same finding using Claude Code, P2Patch raises held-out fix-closed coverage from 78.1% to 84.7% and beyond-upstream coverage from 42.1% to 62.5%. Under the same held-out oracles, it also certifies more attack paths as blocked than recent state-of-the-art agentic repair systems on overlapping subjects. These results show that adversarial testing and independent security oracles expose repair gaps that one-shot validation misses.

A72026-10-14cs.AI

ActionOPD: Distilling Decisions, Not Tokens

Huining Yuan, Xiangmin Yi, Zelai Xu, Zixu Wang, Xinlei Chen, Xiao-Ping Zhang, Yu Wang, Chao Yu

Distilling capable language-model agents into smaller models requires aligning the student's behavior with the teacher's. This raises a fundamental question: at which states and at what granularity should this alignment be performed? Offline imitation provides dense but token-specific supervision and can suffer from state-distribution mismatch. Token-level on-policy distillation (OPD) addresses this mismatch but typically requires teacher token probabilities and compatible tokenization. We argue that semantic actions provide a natural unit of supervision for agent distillation. We introduce ActionOPD, an action-level on-policy distillation method motivated by maximizing expected teacher-action likelihood at student-encountered states. At each such state, ActionOPD samples actions from both the student and a black-box teacher, rewarding student actions according to their semantic agreement with the sampled teacher actions. Through policy optimization, ActionOPD aligns the student's action-level behavior with the teacher's decisions without requiring reproduction of teacher-written reasoning or access to teacher token probabilities. Training Qwen3-4B and Qwen3-8B with two teachers on diverse, curated tool-use tasks yields average gains of 3.3-6.9 percentage points across four out-of-distribution benchmarks. ActionOPD scales favorably with student size and teacher capability, without requiring the teacher and student to share a model family or tokenizer. On τ^2-bench, ActionOPD achieves competitive performance against offline distillation and token-level OPD. We further explore Self-ActionOPD with a same-model teacher and privileged context. These results establish ActionOPD as a promising approach to agent distillation.

A82026-10-14cs.LG

Function-Space Temporal Self-Distillation for Policy Optimization

Aryan Das, Arnab Bhattacharya, Vinay Kumar Verma

Reinforcement Learning with Verifiable Rewards (RLVR) improves reasoning in language models, but scalar outcomes provide coarse supervision over response trajectories. Self-distillation offers denser token-level guidance by comparing an unassisted policy with a feedback-informed self-teacher, yet the teacher evolves with the student, creating unstable targets and motivating additional regularization. We introduce Function-Space Policy Optimization (FSPO), a single-policy framework that combines current privileged adaptation with function-space temporal consolidation. FSPO stores sparse historical next-token distributions as fixed temporal targets, providing temporal separation without maintaining a persistent teacher model. It further introduces privileged disagreement correction to explicitly suppress locally overvalued actions, reliability-adaptive grounding that strengthens reference supervision when self-generated evidence repeatedly fails, and quality and coverage-aware temporal memory that retains verified behaviors and improves privileged demonstrations. Across reasoning and tool-use benchmarks, FSPO achieves the highest average accuracy under a 5-hour training budget, improving over the strongest baseline by approximately 4% points on average while reducing peak memory from about 70 GB to 56 GB on Olmo3-7B-Instruct.

Foundational papers

  1. 2023
    Toolformer: Language models can teach themselves to use tools7,859 citations
  2. 2024
    Toolllm: Facilitating large language models to master 16000+ real-world apis2,608 citations
  3. 2024
    Large language models as tool makers406 citations
  4. 2024
    Hammer: Robust function-calling for on-device language models via function masking59 citations
  5. 2025
    Toolrm: Outcome reward models for tool-calling large language models17 citations
  6. 2026
    Lost in execution: On the multilingual robustness of tool calling in large language models14 citations
  7. 2026-10
    Latest on arXiv: MLExpert: Towards a Generalist Agentic Model for Autonomous Machine Learning Engineeringcs.AI
Citations per paperCitation counts via the AISA Scholar API
citations: R1 The berkeley function… 833, R2 Tool learning with la… 519, R3 Toolllm: Facilitating… 2,608, R4 Toolformer: Language … 7,859, R5 Large language models… 406, R6 Function calling in l… 33, R7 Nexusraven: a commerc… 63, R8 Toolrm: Outcome rewar… 17, R9 Hammer: Robust functi… 59, R10 Lost in execution: On… 14R1 The berkeley function…833R2 Tool learning with la…519R3 Toolllm: Facilitating…2,608R4 Toolformer: Language …7,859R5 Large language models…406R6 Function calling in l…33R7 Nexusraven: a commerc…63R8 Toolrm: Outcome rewar…17R9 Hammer: Robust functi…59R10 Lost in execution: On…14
R2519 citations

Tool learning with large language models: A survey

C Qu, S Dai, X Wei, H Cai, S Wang, D Yin, J Xu et al.

… The overall workflow for tool learning with large language models. The left part illustrates the four stages of tool learning: task planning, tool selection, tool calling, and response …

R420237,859 citations

Toolformer: Language models can teach themselves to use tools

T Schick, J Dwivedi-Yu, R Dessì et al.

… models excel. In this paper, we show that LMs can teach themselves to use external tools via … We introduce Toolformer, a model trained to decide which APIs to call, when to call them, …

R52024406 citations

Large language models as tool makers

T Cai, X Wang, T Ma, X Chen et al.

… convert natural language questions into function calls, and … language model used in the tool-making stage (See Table 3). Generally, we found that a more powerful and expensive model …

How this page is made. Foundational papers and citation counts come from academic search via the AISA Scholar API, with abstracts read from each paper’s page; recent preprints and monthly volumes come from the public arXiv API (metadata CC0). The digest is written by a language model from those abstracts only — every theme links the papers it draws on, and figures not in the inputs are removed.

Sources & citations

  1. The berkeley function calling leaderboard (bfcl): From tool use to agentic evaluation of large language models 833 citations
  2. Tool learning with large language models: A survey 519 citations
  3. Toolllm: Facilitating large language models to master 16000+ real-world apis 2,608 citations
  4. Toolformer: Language models can teach themselves to use tools 7,859 citations
  5. Large language models as tool makers 406 citations
  6. Function calling in large language models: Industrial practices, challenges, and future directions 33 citations
  7. Nexusraven: a commercially-permissive language model for function calling 63 citations
  8. Toolrm: Outcome reward models for tool-calling large language models 17 citations
  9. Hammer: Robust function-calling for on-device language models via function masking 59 citations
  10. Lost in execution: On the multilingual robustness of tool calling in large language models 14 citations
  11. MLExpert: Towards a Generalist Agentic Model for Autonomous Machine Learning Engineering arXiv · 2026-10-15
  12. Learning to Steer LLM Agents and Multi-Agent Systems arXiv · 2026-10-15
  13. PROBE: Learning to Audit Policy Compliance in Tool-Using LLM Agents arXiv · 2026-10-15
  14. RSIFlow: Adaptive Self-Improvement Workflows with an Evolving Meta-Agent arXiv · 2026-10-15
  15. Rethinking the Role of Tool Graph Memory in Multi-Turn Tool Use arXiv · 2026-10-15
  16. P2Patch: Exploit-Guided Vulnerability Repair arXiv · 2026-10-14
  17. ActionOPD: Distilling Decisions, Not Tokens arXiv · 2026-10-14
  18. Function-Space Temporal Self-Distillation for Policy Optimization arXiv · 2026-10-14

Related