REVIEW 23 cited by
Acting Less is Reasoning More! Teaching Model to Act Efficiently
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
Acting Less is Reasoning More! Teaching Model to Act Efficiently
read the original abstract
Tool-integrated reasoning (TIR) augments large language models (LLMs) with the ability to invoke external tools during long-form reasoning, such as search engines and code interpreters, to solve tasks beyond the capabilities of internal reasoning. While reinforcement learning (RL) has shown promise in training such agents, most of existing approaches typically optimize only for final correctness without considering the efficiency or necessity of external tool use. This often leads to excessive tool calling, incurring high computational costs and hindering the development of internal reasoning capabilities - a phenomenon known as \textit{cognitive offloading}. To this end, we propose Optimal Tool Call-controlled Policy Optimization (OTC-PO), a simple yet effective RL-based framework that encourages models to produce accurate answers with minimal tool calls. Our method introduces a tool-integrated reward that jointly considers answer correctness and corresponding tool use behavior of model to reach that answer. To validate the effectiveness, we introduce the metric of \textit{tool productivity}, defined as the ratio between the number of correct answers and the total number of tool calls across all test cases. This metric reflects how efficiently tool usage contributes to successful task completion, with higher values indicating smarter and more autonomous reasoning. We instantiate this framework within both Proximal Policy Optimization (PPO) and Group Relative Preference Optimization (GRPO), resulting in OTC-PPO and OTC-GRPO. Experiments with Qwen-2.5 and Qwen-Math across multiple QA benchmarks show that our approach reduces tool calls by up to 68.3\% and improves tool productivity by up to 215.4\%, while maintaining comparable answer accuracy.
Forward citations
Cited by 23 Pith papers
-
The Piggyback Hypothesis of Generalization: Explaining and Mitigating Emergent Misalignment
Shared chat-template tokens piggyback narrow finetuning behaviors onto out-of-domain queries; regularizing their KV states (TReFT) reduces emergent misalignment and other off-topic generalization.
-
Train the Agent, Not the Expert: Learning to Harness Heterogeneous Experts for Multi-Turn Visual Reasoning
VisHarness learns a reinforcement-learned policy to harness specialized visual experts via multi-turn interactions and dynamic visual memory archiving, outperforming general models on four visual reasoning benchmarks.
-
Draw2Think: Harnessing Geometry Reasoning through Constraint Engine Interaction
Draw2Think recasts geometric reasoning as agentic interaction with a constraint engine, achieving 95.9% predicate-level construction fidelity and up to 16.4% accuracy gains on solid geometry tasks.
-
Group-in-Group Policy Optimization for LLM Agent Training
GiGPO adds a hierarchical grouping mechanism to group-based RL so that LLM agents receive both global trajectory and local step-level credit signals, yielding >12% gains on ALFWorld and >9% on WebShop over GRPO while ...
-
Diagnosing Search Behavior and Failure Modes in Long-Horizon Search Agents
For six deep search agents on BrowseComp-Plus, answer accuracy tracks cumulative retrieval recall, not search effort, and failures split into missing-evidence and evidence-misuse gaps.
-
ToolSciVer: Multimodal Scientific Claim Verification with Visual Tool Augmented Reinforcement Learning
Reinforcement-learned, type-aware visual tool calls improve multimodal scientific claim verification on SCIVER and MuSciClaims.
-
Tool-Adaptive LLM Reranker
Pointwise LLM reranking cast as an agentic MDP with language-preserving warm-up and asymmetric cost-aware RL yields adaptive tool use, SOTA NDCG, and high throughput.
-
The Piggyback Hypothesis of Generalization: Explaining and Mitigating Emergent Misalignment
The Piggyback Hypothesis attributes emergent misalignment to chat-template tokens piggybacking finetuned behavior; Token-Regularized Finetuning (TReFT) mitigates it by regularizing prefix token representations.
-
Agent Explorative Policy Optimization for Multimodal Agentic Reasoning
AXPO addresses the Thinking-Acting Gap in agentic RL training by targeted resampling of tool calls in all-wrong subgroups, delivering +1.8pp gains over GRPO on nine multimodal benchmarks with an 8B model beating a 32B...
-
ToolCUA: Towards Optimal GUI-Tool Path Orchestration for Computer Use Agents
ToolCUA introduces a trajectory scaling pipeline and staged RL to optimize GUI-tool switching, reaching 46.85% accuracy on OSWorld-MCP for a 66% relative gain over baseline.
-
TIDE-Bench: Task-Aware and Diagnostic Evaluation of Tool-Integrated Reasoning
TIDE-Bench is a new benchmark for tool-integrated reasoning that combines diverse tasks, multi-aspect metrics covering answer quality, process reliability, efficiency and cost, plus filtered challenging test sets.
-
Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence
Agent-World autonomously synthesizes verifiable real-world tasks and uses continuous self-evolution to train 8B and 14B agents that outperform proprietary models on 23 benchmarks.
-
TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework
TeaRAG shows that hybrid chunk+triplet retrieval with Personalized PageRank and an iterative process-aware DPO reward keeps QA accuracy while cutting reasoning tokens by roughly 60%.
-
MASH: Modeling Abstention via Selective Help-Seeking
MASH uses RL with a pay-per-search reward to make LLMs seek external help only when needed, improving multi-hop QA accuracy by 7.6% and enabling competitive abstention without pre-defined knowledge boundaries.
-
Efficient and Transferable Agentic Knowledge Graph RAG via Reinforcement Learning
KG-R1 trains a single RL agent to retrieve from and reason over knowledge graphs in one loop, achieving higher accuracy with fewer tokens than multi-module baselines and transferring to unseen graphs.
-
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
Survey that defines agentic RL for LLMs via POMDPs, introduces a taxonomy of planning/tool-use/memory/reasoning capabilities and domains, and compiles open environments from over 500 papers.
-
WebThinker: Empowering Large Reasoning Models with Deep Research Capability
WebThinker equips large reasoning models with autonomous web exploration and interleaved reasoning-drafting via a Deep Web Explorer and RL-based DPO training, yielding gains on GPQA, GAIA, and report-generation benchmarks.
-
Contrastive Reinforced Policy Optimization via Privileged Self-Distillation
CRPO turns on-policy self-distillation into group-wise contrastive learning gated by student–teacher entropy gaps, improving multi-turn agentic LLM post-training over GRPO, ARPO, and OPSD.
-
Einstein World Models
Einstein World Models integrate visual rollouts from a callable world-module into LLM reasoning traces to support complex thought beyond language.
-
IAPO: Input Attribution-Aware Policy Optimization for Tool Use in Small Multimodal Agents
IAPO is an RL method that aligns model input attributions with a teacher to improve tool-calling in multimodal SLMs, reporting 3% average VQA accuracy gains on Qwen2.5-VL-3B across six tests.
-
E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning
E3-TIR integrates expert prefixes, guided branches, and self-exploration via mix policy optimization to deliver 6% better tool-use performance with under 10% of the usual synthetic data and 1.46x ROI.
-
UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning
UI-TARS-2 reaches 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld while attaining 59.8 mean normalized score on a 15-game suite through multi-turn RL and scalable data generation.
-
Toward Efficient Agents: Memory, Tool learning, and Planning
A survey that organizes efficiency techniques for LLM agents into memory, tool learning, and planning, and consolidates benchmarks and metrics for measuring cost-performance trade-offs.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.