ABSeeker: Training Long-Horizon Search Agents via Answer-Backtracked Credit Assignment

• URL: http://arxiv.org/abs/2608.05102v1
• Date: Wed, 05 Aug 2026 17:41:31 GMT
• Title: ABSeeker: Training Long-Horizon Search Agents via Answer-Backtracked Credit Assignment
• Abstract summary: Long-horizon search agents must make multiple sequential actions (steps) to search, retrieve, verify, and integrate evidence to reach a final answer.<n>We propose Answer-Backtracked Credit Assignment (ABC), a fine-grained credit assignment framework for training long-horizon search agents.<n>ABC converts sparse trajectory-level outcomes into dense step-level supervision that rewards useful actions (even in failed trajectories) while suppressing erroneous or redundant actions.
• Score: 66.45065414461274
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Long-horizon search agents must make multiple sequential actions (steps) to search, retrieve, verify, and integrate evidence to reach a final answer. However, existing methods for training these agents typically treat all steps within a trajectory uniformly during both supervised fine-tuning (SFT) and reinforcement learning (RL), failing to distinguish useful actions from erroneous or redundant ones. In this paper, we propose Answer-Backtracked Credit Assignment (ABC), a fine-grained credit assignment framework for training long-horizon search agents by converting sparse trajectory-level outcomes into dense step-level supervision that rewards useful actions (even in failed trajectories) while suppressing erroneous or redundant actions. Specifically, given a potentially obscure query and its corresponding ground-truth answer, ABC first performs Answer-Backtracked Clue Recovery, which traces back from the answer to recover intermediate clues required to solve the question. It then applies Clue-Anchored Step Scoring to evaluate each search step against these clues, converting sparse binary outcome supervision into dense step-level rewards. Based on these rewards, we develop ABC-SFT, which reweights the loss of each turn, and ABC-GRPO, which uses the step-level scores as rewards in GRPO. Building on this framework, we train ABSeeker based on Qwen3.5-4B with only 8.5k examples. ABSeeker achieves 37.3% on BrowseComp and 39.1% on BrowseComp-ZH. With context management, the scores further improve to 55.3% and 52.9%, respectively, significantly outperforming same-scale (4B) agents and even matching the performance of larger ones (approximately 30B). These results demonstrate the effectiveness of answer-backtracked step-level credit assignment for training long-horizon search agents.

Related papers

• TRCA: Transition-wise Rubric Credit Assignment for Long-horizon LLM Agents [22.01777615106231]
  Long-horizon large language model (LLM) agents are typically optimized with sparse terminal outcomes.<n>We propose Transition-wise Credit Assignment, which derives step-level supervision directly from action-induced transitions.<n>Experiments on ALFWorld, WebShop, and seven search-augmented question-answering benchmarks show consistent improvements over the evaluated baselines.
  arXiv  Detail & Related papers  (2026-08-17T06:19:12Z)
• SearchAuditor: Auditing and Attributing Failures in Long-Horizon Search Agents [56.400332595452966]
  We introduce SearchAuditBench, a benchmark that evaluates whether auditors can localize, attribute, and repair search-agent failures.<n>SearchAuditBench comprises 1,243 failed trajectories, averaging 73.1 messages and 65.1K tokens, collected from eight open-weight models on five deep-search benchmarks.<n>We propose SearchAuditor, a multi-perspective auditing framework that effectively localizes, attributes, and repairs search-agent failures.
  arXiv  Detail & Related papers  (2026-08-05T09:09:00Z)
• BiCAA: Bidirectional Credit Assignment for Search-Augmented Agent [7.0011576165695]
  BiCAA builds bidirectional process rewards by fusing two complementary signals: forward solvability gain and hindsight success criticality.<n> Experiments on search-augmented QA benchmarks show that BiCAA stabilizes policy optimization, reduces redundant search behavior, and achieves competitive performance.
  arXiv  Detail & Related papers  (2026-08-02T15:41:57Z)
• TRACE: Turn-level Reward Assignment via Credit Estimation for Long-Horizon Agents [51.523913302114266]
  TRACE (Turn-level Reward Assignment via Credit Estimation) is a dense credit-assignment method for agentic reinforcement learning.<n>It represents rollouts as state transitions at tool-call boundaries, obtains gold-answer log-probabilities from a frozen reference model, transforms them into log-ratio state values, and derives per-action rewards as Temporal-Difference changes in those values.<n>It substantially improves base-model tool-use ability using pure RL, without a cold-start supervised fine-tuning stage, an agentic mid-training stage, or training on live-web data.
  arXiv  Detail & Related papers  (2026-07-15T16:16:42Z)
• ARBOR: Online Process Rewards via a Reusable Rubric Buffer for Search Agents [48.80766702702854]
  LLM-based search agents are trained predominantly with outcome-only reward, leaving the search process itself unsupervised.<n>This signal degenerates on outcome-homogeneous groups where all sampled trajectories share the same correctness, yielding zero within-group advantage and no gradient.<n>We propose ARBOR (Adaptive Buffer for Online Reward), a reusable process-reward framework that maintains a rubric memory shared across queries.
  arXiv  Detail & Related papers  (2026-06-02T06:58:54Z)
• Cycle-Consistent Search: Question Reconstructability as a Proxy Reward for Search Agent Training [80.20022221643414]
  Cycle-Consistent Search is a gold-supervision-free framework for training search agents.<n>We show that CCS achieves performance comparable to supervised baselines.<n>These results suggest that CCS provides a scalable training paradigm for training search agents in settings where gold supervision is unavailable.
  arXiv  Detail & Related papers  (2026-04-14T17:00:18Z)
• Training Multi-Turn Search Agent via Contrastive Dynamic Branch Sampling [29.182538022605627]
  Branching Relative Policy Optimization (BranPO) is a value-free method that provides step-level contrastive supervision without dense rewards.<n>BranPO truncates trajectories near the tail and resamples alternative continuations to construct contrastive suffixes over shared prefixes.<n>To further boost efficiency and stabilize training, we introduce difficulty-aware branch sampling to adapt branching frequency across tasks, and redundant step masking to suppress uninformative actions.
  arXiv  Detail & Related papers  (2026-02-03T16:43:09Z)
• RE-TRAC: REcursive TRAjectory Compression for Deep Search Agents [144.5598958575922]
  Re-TRAC is an agentic framework that performs cross-trajectory exploration.<n>We show that Re-TRAC consistently outperforms ReAct by 15-20% on BrowseComp with frontier LLMs.
  arXiv  Detail & Related papers  (2026-02-02T18:58:07Z)
• Beyond Outcome Reward: Decoupling Search and Answering Improves LLM Agents [19.31471304268234]
  We introduce DeSA (Decoupling Search-and-Answering), a simple two-stage training framework that explicitly separates search optimization from answer generation.<n>Across seven QA benchmarks, DeSA-trained agents consistently improve search behaviors, delivering substantially higher search recall and answer accuracy than outcome-only baselines.
  arXiv  Detail & Related papers  (2025-10-06T11:09:45Z)
• RLVMR: Reinforcement Learning with Verifiable Meta-Reasoning Rewards for Robust Long-Horizon Agents [43.806220882212386]
  RLVMR integrates dense, process-level supervision into end-to-end RL by rewarding verifiable, meta-reasoning behaviors.<n>On the challenging ALFWorld and ScienceWorld benchmarks, RLVMR achieves new state-of-the-art results.
  arXiv  Detail & Related papers  (2025-07-30T17:00:48Z)
• RRO: LLM Agent Optimization Through Rising Reward Trajectories [52.579992804584464]
  Large language models (LLMs) have exhibited extraordinary performance in a variety of tasks.<n>In practice, agents sensitive to the outcome of certain key steps which makes them likely to fail the task.<n>We propose Reward Rising Optimization (RRO) to mitigate this issue.
  arXiv  Detail & Related papers  (2025-05-27T05:27:54Z)

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.