Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs

• URL: http://arxiv.org/abs/2509.25779v2
• Date: Wed, 01 Oct 2025 20:23:23 GMT
• Title: Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs
• Abstract summary: We investigated agentic RL with large language models on the textscTravelPlanner benchmark.<n>Our approach, textscNatural-R1, achieved a textbf56.9% final-pass rate with only 180 training queries.
• Score: 10.044336754070203
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We investigated Agentic RL with large language models on the \textsc{TravelPlanner} benchmark. Our approach, \textsc{Planner-R1}, achieved a \textbf{56.9\%} final-pass rate with only 180 training queries, a $2.7\times$ improvement over GPT-5's $21.2\%$ baseline and the strongest agentic result on the public leaderboard. A central finding was that smaller models (8B) were highly responsive to reward shaping: with dense process-level signals, they reached competitive performance while being $3.5\times$ more compute-efficient and $1.5\times$ more memory-efficient than 32B models. Larger models were more robust under sparse rewards but exhibited smaller relative gains from shaping and higher variance across runs. While curriculum learning offered no significant benefit, shaped rewards consistently amplified learning dynamics, making 8B models the most efficient setting for agentic RL. Crucially, these gains did not come at the cost of overfitting: fine-tuned models mostly maintained or exceeded baseline performance on out-of-domain tasks, including \textsc{Multi-IF}, \textsc{NaturalPlan}, and $\tau$-\textsc{Bench}. These results establish reward shaping as a decisive lever for scaling agentic RL, highlight the competitive strength of smaller models, and demonstrate that efficiency can be achieved without sacrificing generalization.

Related papers

• Models for minimalist RAG: B1ade 335M Embedding and 1B Parameter Small Language Models [2.1665689529884697]
  We present B1ade, an efficient RAG architecture comprising two purpose-built components: a compact embedding model and a purpose-built SLM.<n>B1ade-embed, a 335M parameter retrieval model constructed via parameter-free fusion of five pretrained encoders achieves top MTEB scores among sub-500M models with zero additional training.<n>B1ade-1B cites retrieved passages in 42.4% of responses, exceeding the attribution rate of its training distribution by 5.5 percentage points.
  arXiv  Detail & Related papers  (2026-07-29T22:41:25Z)
• MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in Compact Models [31.136687654577475]
  We introduce MADA-RL, a post-training framework that specializes compact models into generator and critic roles.<n>Our central contribution is a counterfactual critic advantage that redefines the critic's advantage as its reward minus the generator ensemble's per-instance accuracy.
  arXiv  Detail & Related papers  (2026-07-20T14:38:00Z)
• Mach-Mind-4-Flash Technical Report [0.0]
  Mach-Mind-4-Flash is a Mixture-of-Experts (MoE) agentic model with 3B activated parameters.<n>Post-training optimization alone without scaling pre-training compute, the model achieves performance on par with or surpassing that of 100B- parameter-class models.
  arXiv  Detail & Related papers  (2026-07-10T12:57:08Z)
• PRISM: Demystifying Retention and Interaction in Mid-Training [20.198164159173647]
  PRISM is a comprehensive empirical study of mid-training design choices for large language models.<n>We show that mid-training on approximately 27B high-quality tokens yields consistent gains of +15 to +40 points on math, +5 to +12 points on code, and +6 to +13 points on science benchmarks while preserving general performance.
  arXiv  Detail & Related papers  (2026-03-17T19:04:33Z)
• Scaling Reward Modeling without Human Supervision [52.10639750993359]
  We operationalize reward-based scaling by preference learning over document prefixes and suffixes drawn from large-scale web corpora.<n>Despite using no human annotations, training on 11M tokens of math-focused web data yields steady gains on RewardBench v1 and v2.<n>Across models, our method improves RewardBench v2 accuracy by up to +7.7 points on average, with gains of up to +16.1 on in-domain math subsets.
  arXiv  Detail & Related papers  (2026-02-11T04:41:53Z)
• Distributional Clarity: The Hidden Driver of RL-Friendliness in Large Language Models [50.99097734404912]
  We show that RL-friendly models exhibit intra-class compactness and inter-class separation in their probability assignments to correct vs. incorrect responses.<n>Experiments across six mathematical benchmarks show consistent improvements across all model families, with gains up to 5.9 points on AIME24.
  arXiv  Detail & Related papers  (2026-01-11T13:34:44Z)
• Miner:Mining Intrinsic Mastery for Data-Efficient RL in Large Reasoning Models [40.61814017829362]
  Current critic-free RL methods for large reasoning models suffer from severe inefficiency when training on positive homogeneous prompts.<n>We introduce a radically simple yet powerful solution to ulineMine ulineintrinsic mastulineery (Miner)<n>Miner repurposes the policy's intrinsic uncertainty as a self-supervised reward signal.
  arXiv  Detail & Related papers  (2026-01-08T08:52:37Z)
• Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model [100.86587937568832]
  Ring-1T is the first open-source, state-of-the-art thinking model with a trillion-scale parameter.<n>It features 1 trillion total parameters and activates approximately 50 billion per token.
  arXiv  Detail & Related papers  (2025-10-21T17:46:14Z)
• GRPO-$λ$: Credit Assignment improves LLM Reasoning [35.452488047246646]
  We present GRPO-$lambda$, a novel extension to GRPO that enhances credit assignment in RL finetuning of LLMs for complex reasoning tasks.<n>We compare GRPO-$lambda$ against GRPO by training models from 1.5B to 7B parameters on $4$ different math reasoning datasets.<n>With GRPO-$lambda$, the resulting average performance on AIME24, Math500, OlympiadMath, MinervaMath, and AMC improves over GRPO by over $3$ points and a $4.5$ points improvement on the 7B model.
  arXiv  Detail & Related papers  (2025-09-30T19:11:10Z)
• Intention-Conditioned Flow Occupancy Models [69.79049994662591]
  Large-scale pre-training has fundamentally changed how machine learning research is done today.<n>Applying this same framework to reinforcement learning is appealing because it offers compelling avenues for addressing core challenges in RL.<n>Recent advances in generative AI have provided new tools for modeling highly complex distributions.
  arXiv  Detail & Related papers  (2025-06-10T15:27:46Z)
• Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning [93.00629872970364]
  Reinforcement learning (RL) has become the dominant paradigm for improving the performance of language models on complex reasoning tasks.<n>We introduce SPARKLE, a fine-grained analytic framework to dissect the effects of RL across three key dimensions.<n>We study whether difficult problems -- those yielding no RL signals and mixed-quality reasoning traces -- can still be effectively used for training.
  arXiv  Detail & Related papers  (2025-06-05T07:53:59Z)
• Skywork Open Reasoner 1 Technical Report [51.403686909760914]
  We present Skywork-OR1, an effective and scalable reinforcement learning (RL) implementation for long Chain-of-Thought (CoT) models.<n>Building on the DeepSeek-R1-Distill model series, our RL approach achieves notable performance gains.<n>Our Skywork-OR1-32B model surpasses both DeepSeek-R1 and Qwen3-32B on the AIME24 and AIME25 benchmarks.
  arXiv  Detail & Related papers  (2025-05-28T12:56:04Z)
• Adapt-Pruner: Adaptive Structural Pruning for Efficient Small Language Model Training [27.857935426067076]
  Small language models (SLMs) have attracted considerable attention due to their broad range of applications in edge devices.<n>To obtain SLMs with strong performance, conventional approaches either pre-train the models from scratch, which incurs substantial computational costs, or compress/prune existing large language models (LLMs), which results in performance drops and falls short in comparison to pre-training.<n>We found 1) layer-wise adaptive pruning (Adapt-Pruner) is extremely effective in LLMs and yields significant improvements over existing pruning techniques, 2) adaptive pruning equipped with further training leads to models comparable to those pre-training from scratch
  arXiv  Detail & Related papers  (2025-02-05T18:57:40Z)
• Does RLHF Scale? Exploring the Impacts From Data, Model, and Method [83.53178716807776]
  This study explores the scaling properties of Reinforcement Learning from Human Feedback in Large Language Models.<n>We analyze key components in the RLHF framework--model size, data composition, and inference budget--and their impacts on performance.
  arXiv  Detail & Related papers  (2024-12-08T17:19:48Z)
• RLx2: Training a Sparse Deep Reinforcement Learning Model from Scratch [23.104546205134103]
  Training deep reinforcement learning (DRL) models usually requires high costs. compressing DRL models possesses immense potential for training acceleration and model deployment. We propose a novel sparse DRL training framework, "the textbfRigged textbfReinforcement textbfLearning textbfLottery" (RLx2)
  arXiv  Detail & Related papers  (2022-05-30T12:18:43Z)
• Exploring Sparse Expert Models and Beyond [51.90860155810848]
  Mixture-of-Experts (MoE) models can achieve promising results with outrageous large amount of parameters but constant computation cost. We propose a simple method called expert prototyping that splits experts into different prototypes and applies $k$ top-$1$ routing. This strategy improves the model quality but maintains constant computational costs, and our further exploration on extremely large-scale models reflects that it is more effective in training larger models.
  arXiv  Detail & Related papers  (2021-05-31T16:12:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.