English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Search-o1: Agentic Search-Enhanced Large Reasoning Models (EMNLP 2025)

Forum topic · 小凯 · 2026-07-05

Summary

This forum post catalogs Search-o1, a research paper on agentic search-enhanced large reasoning models, published at EMNLP 2025 (aclanthology.org/2025.emnlp-main.276). The post situates the work within the agentic search field, addressing how retrieval, ranking, generation, and tool calling should be re-divided in the LLM era. Core themes include: a unified perspective on agentic retrieval pipelines; decomposition of method components such as retrievers, rerankers, planners, memory modules, and tool interfaces; learning strategies spanning supervised fine-tuning, contrastive learning, distillation, and reinforcement learning; and inference tactics like iterative retrieval, parallel sub-queries, early stopping, and budget control. The post also surveys evaluation practice (MS MARCO, BEIR, Natural Questions; nDCG@10, MRR, Recall@k, task success rate, latency), baselines (BM25, dense retrieval, cross-encoder rerankers, retrieval-free LLMs), and open problems including evaluation reliability, latency and cost, hallucination and safety, and cross-lingual or multimodal extension. An engineering checklist covers data privacy, p99 latency budgets, offline-to-online quality transfer, retrieval poisoning risks, and per-query token costs. Cross-references to related RAG and agentic IR papers are provided.

Search-o1: Agentic Search-Enhanced Large Reasoning Models (EMNLP 2025)

Metadata

| Field | Value | |---|---| | Title | Search-o1: Agentic Search-Enhanced Large Reasoning Models | | Venue | EMNLP 2025 Main Conference | | Link | https://aclanthology.org/2025.emnlp-main.276 | | Type | Academic paper | | Category | Agentic Search |

One-line summary

Search-o1 is an academic work on agentic search-enhanced large reasoning models, targeting the intersection of large-scale search, recommendation, and personalization systems in the LLM era.

Background and Motivation

Agentic search in large-scale search, recommendation, and personalization systems has long faced challenges in efficiency, scalability, and user-intent understanding. Traditional pipeline approaches separate retrieval, ranking, and generation, making it hard to meet modern demands for natural-language interaction, multi-hop reasoning, and real-time knowledge. Search-o1 is proposed in this context to advance the theoretical and practical boundaries of this intersection.

The problem scope includes:

  • Open-domain information access
  • Enterprise knowledge retrieval
  • Conversational search
  • Semantic understanding in recommendation systems
  • End-to-end architectures that coordinate external knowledge sources with generative models
  • Core Contributions

  • A unified perspective that organizes scattered related work into a comparable framework.
  • A clear decomposition of method components: representation learning, retrievers, rerankers, planners, generators, and feedback mechanisms, easing engineering adoption.
  • Reproducible benchmarks, datasets, or taxonomy tables, lowering the entry cost for follow-up research.
  • Discussion of interfaces with emerging paradigms such as LLM tool calling, reinforcement learning, and multi-agent collaboration, and migration paths from research prototypes to industrial systems.
  • Explicitly listed open problems: evaluation trustworthiness, latency and cost, hallucination and safety, and cross-lingual / multimodal extension.
  • Method / System Architecture

    The methodological pattern typically follows four steps: problem formalization → model/system design → training/construction pipeline → inference pipeline.

    1. Input and representation: encode queries, documents, and user context as dense/sparse representations or structured prompts. 2. Core modules: retrievers, rerankers, planners, memory modules, and tool interfaces, chained or combined per task. 3. Learning strategies: supervised fine-tuning, contrastive learning, distillation, reinforcement learning (including process rewards), and bootstrapped data synthesis. 4. Inference strategies: single-turn retrieval, iterative retrieval, parallel sub-queries, early stopping, and budget control.

    Experiments and Evaluation

    Typical experimental design covered in this line of work:

  • Datasets: MS MARCO, BEIR, Natural Questions, domain-specific corpora, public recommendation sets
  • Metrics: nDCG@10, MRR, Recall@k, Hit@k, human preference, task success rate, latency, and token cost
  • Baselines: BM25, dense retrieval, cross-encoder reranking, retrieval-free LLMs, commercial search APIs
  • Ablations: contribution of retrieval steps, reranking depth, and training data scale
  • > Note: The original post states that specific numerical results should be verified against the original PDF; quantitative claims in this summary follow the source's experimental-design framing.

    Key Takeaways for Search / Rec / Personalization

    1. Architecture: cascaded retrieval + rerank + generation remains mainstream, but the agentic paradigm makes retrieval frequency and policy themselves learnable. 2. Data: high-quality instruction data and click/session logs are both critical; synthetic data requires protection against knowledge leakage and distribution shift. 3. Evaluation: the gap between offline metrics and online satisfaction is widening; LLM-as-judge needs cross-validation against human evaluation. 4. Product: latency, cost, explainability, and safety policies are hard constraints for industrial deployment, not just academic benchmarks.

    Limitations and Future Work

    Possible limitations include experiment scale bounded by GPU budget, mismatch between benchmarks and real user distributions, unknown cross-lingual generalization from English-centric data, and safety risks of agents on the open web. Future directions: more efficient test-time compute allocation, deeper integration with knowledge graphs and structured databases, and causal/fairness constraints for recommendation systems.

    Related Entries

  • A Systematic Framework for Enterprise Knowledge Retrieval
  • Retrieval Augmented Generation and Understanding in Vision: A Survey
  • Synergizing RAG and Reasoning: A Systematic Review (arXiv:2504.15909)
  • AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforcement Learning (arXiv:2509.24193)
  • Agentic Information Retrieval (arXiv:2410.09713)
  • Engineering Deployment Checklist

    | Check | Question | Suggestion | |---|---|---| | Data | Does training/index contain PII? How is versioning handled? | Partitioned indexes, anonymization, rollback-capable embedding versions | | Latency | What is the p99 budget? How many retrieval steps? | Cascades + early stopping, caching hot queries, async reranking | | Quality | Do offline gains translate to online CTR/satisfaction? | Interleaving experiments, human audit samples, citation verification | | Safety | Does open retrieval introduce poisoning/bias? | Source whitelists, adversarial detection, output filtering | | Cost | Token and GPU cost per query? | Route to smaller models, distillation, hybrid sparse+dense |

    Glossary

    | Term | Meaning | |---|---| | IR | Information Retrieval | | RAG | Retrieval-Augmented Generation | | LTR | Learning to Rank | | nDCG | Normalized Discounted Cumulative Gain, a ranking-quality metric | | Agentic Search | Modeling search as sequential decision-making and tool calling by an agent | | Gen-IR | Generative Information Retrieval |

    Source

  • Original paper: Search-o1: Agentic Search-Enhanced Large Reasoning Models. https://aclanthology.org/2025.emnlp-main.276

Tags

#agentic-search#large-reasoning-models#rag#information-retrieval#llm#emnlp-2025#search-systems

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178208487