REVIEW 27 cited by
SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning
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
SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning
read the original abstract
Large Language Models (LLMs) can significantly improve their reasoning capabilities by interacting with external tools, a paradigm known as Tool-Integrated Reasoning (TIR). However, extending TIR to multi-turn scenarios using Reinforcement Learning (RL) is often hindered by training instability and performance collapse. We identify that such instability is primarily caused by a distributional drift from external tool feedback, leading to the generation of low-probability tokens. This issue compounds over successive turns, causing catastrophic gradient norm explosions that derail the training process. To address this challenge, we introduce SimpleTIR , a plug-and-play algorithm that stabilizes multi-turn TIR training. Its core strategy is to identify and filter out trajectories containing void turns, i.e., turns that yield neither a code block nor a final answer. By removing these problematic trajectories from the policy update, SimpleTIR effectively blocks the harmful, high-magnitude gradients, thus stabilizing the learning dynamics. Extensive experiments show that SimpleTIR achieves state-of-the-art performance on challenging math reasoning benchmarks, notably elevating the AIME24 score from a text-only baseline of 22.1 to 50.5 when starting from the Qwen2.5-7B base model. Furthermore, by avoiding the constraints of supervised fine-tuning, SimpleTIR encourages the model to discover diverse and sophisticated reasoning patterns, such as self-correction and cross-validation.
Forward citations
Cited by 27 Pith papers
-
Extrapolative Weight Averaging Reveals Correctness-Efficiency Frontiers in Code RL
Extrapolative weight averaging of RL checkpoints trained under nested unit-test coverage extends a correctness-efficiency frontier and boosts ensemble pass rates in code generation across model scales and inference modes.
-
The Cancellation Hypothesis in Critic-Free RL: From Outcome Rewards to Token Credits
The cancellation hypothesis shows how rollout-level rewards produce token-level credit assignment in critic-free RL through cancellation of opposing signals on shared tokens, with empirical support and batching interv...
-
Rethinking Importance Sampling in LLM Policy Optimization: A Cumulative Token Perspective
The cumulative token IS ratio gives unbiased prefix correction and lower variance than full-sequence ratios for token-level gradients in LLM policy optimization, enabling CTPO to outperform GRPO and GSPO baselines on ...
-
Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning
VideoDR is a new benchmark for open-web video deep research that tests multimodal models on cross-frame visual anchor extraction, interactive retrieval, and multi-hop reasoning over joint video-web evidence.
-
SCRIBE: Structured Mid-Level Supervision for Tool-Using Language Models
SCRIBE introduces skill-conditioned rewards with intermediate behavioral evaluation to reduce noise in training tool-augmented agents, raising AIME25 accuracy from 43.3% to 63.3% on a Qwen3-4B model.
-
Training Multi-Image Vision Agents via End2End Reinforcement Learning
IMAgent trains a multi-image vision agent via pure end-to-end RL with visual reflection tools and a two-layer motion trajectory masking strategy, reaching SOTA on single- and multi-image benchmarks while revealing too...
-
Behavior Leverage Imbalance in Multi-Teacher On-Policy Distillation
Multi-teacher on-policy distillation can cause tool over-calling due to disproportionate signals at mode-entry tokens, and a per-token divergence calibration method called Soft Clamp mitigates this shift.
-
Learning When Not to Act: Mitigating Tool Abuse in Agentic Reinforcement Learning
EAPO learns selective tool use in agentic RL via tool-free trajectories, difficulty-aware reward shaping, and confidence-aware token reweighting, improving accuracy while cutting tool calls versus GRPO on nine reasoni...
-
Efficient Agentic Reasoning Through Self-Regulated Simulative Planning
SR²AM achieves competitive Pass@1 accuracy on diverse tasks with 25.8-95.3% fewer reasoning tokens than much larger models by using self-regulated simulative planning trained via supervised learning and RL.
-
Harnessing LLM Agents with Skill Programs
HASP upgrades textual skills into executable Program Functions that intervene in LLM agent loops at inference, post-training, or self-evolution, delivering 25% gains over ReAct and 30.4% over Search-R1 on reasoning be...
-
PruneTIR: Inference-Time Tool Call Pruning for Effective yet Efficient Tool-Integrated Reasoning
PruneTIR prunes erroneous tool-call trajectories during LLM inference via three trigger-based components to raise Pass@1 accuracy and efficiency while shortening context.
-
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.
-
SOD: Step-wise On-policy Distillation for Small Language Model Agents
SOD reweights on-policy distillation strength step-by-step using divergence to stabilize tool use in small language model agents, yielding up to 20.86% gains and 26.13% on AIME 2025 for a 0.6B model.
-
Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning
A training recipe for tool-integrated reasoning models achieves state-of-the-art open-source results on math benchmarks such as 96.7% and 99.2% on AIME 2025 at 4B and 30B scales by balancing tool-use trajectories and ...
-
T$^2$PO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning
T²PO improves stability and performance in multi-turn agentic RL by using uncertainty dynamics at token and turn levels to guide exploration and avoid wasted rollouts.
-
When to Trust Tools? Adaptive Tool Trust Calibration For Tool-Integrated Math Reasoning
ATTC reduces 'Tool Ignored' errors in tool-integrated reasoning by adaptively trusting tool results according to generated code confidence, yielding 4.1-7.5% gains across models and datasets.
-
AnomalyAgent: Agentic Industrial Anomaly Synthesis via Tool-Augmented Reinforcement Learning
AnomalyAgent uses tool-augmented reinforcement learning with self-reflection to generate realistic industrial anomalies, achieving better metrics than zero-shot methods on MVTec-AD.
-
Walk the Talk: Bridging the Reasoning-Action Gap for Thinking with Images via Multimodal Agentic Policy Optimization
MAPO improves multimodal chain-of-thought reasoning by requiring explicit textual descriptions of visual tool results and using a novel advantage estimator that combines semantic alignment with task rewards.
-
Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It
RL for LLM multi-step tool use collapses from control token probability spikes but interleaving SFT improves stability at the cost of OOD generalization.
-
Latent Visual States for Efficient Multimodal Reasoning
EVA generates adaptive Latent_slot tokens as internal visual thoughts, trained end-to-end with text tokens via D-GSPO on the EVA-230K dataset, claiming performance gains and better inference efficiency.
-
Pushing the Limits of LLM Tool Calling via Experiential Knowledge Integration and Activation
Instance-level experiential knowledge provides strong gains for LLM tool calling, parallel sampling activates it more effectively than deeper reasoning, and RL-based internalization outperforms SFT, yielding the KATE ...
-
Local Guidance, Global Impact: Gaussian-Reshaped Trust Region Unlocks Behavior Transitions
GTR introduces a bounded non-monotonic Gaussian trust region and Mixture Gaussian Anchor to enable effective behavior transitions in non-stationary RL where standard PPO fails.
-
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.
-
MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning
MemOCR renders structured memory as images with adaptive visual density to improve long-horizon reasoning under tight context budgets.
-
Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search
Mini-o3 scales visual search reasoning to tens of interaction turns via a new probe dataset, iterative trajectory collection, and over-turn masking in RL, claiming SOTA performance while training only up to six turns.
-
AIR: Adaptive Interleaved Reasoning with Code in MLLMs
AIR applies RL with a group-constrained reward and custom data pipeline to enable adaptive code-interleaved reasoning in MLLMs, reporting 6.1 pp average benchmark gains.
-
The Agent Use of Agent Beings: Agent Cybernetics Is the Missing Science of Foundation Agents
Agent Cybernetics reframes foundation agent design by adapting classical cybernetics laws into three engineering desiderata for reliable, long-running, self-improving agents.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.