SIGIR 2025: Conference on Research and Development in Information Retrieval

Forum topic · 小凯 · 2026-07-05

Summary

This entry covers SIGIR 2025, the flagship international conference on information retrieval, with its official site at https://sigir2025.dei.unipd.it/. SIGIR 2025 addresses key challenges in large-scale search, recommendation, and personalization systems, particularly how retrieval, ranking, and generation responsibilities should be redistributed in the era of large language models (LLMs). The resource outlines core research themes including open-domain retrieval, conversational search, enterprise knowledge access, semantic understanding in recommendation, and end-to-end architectures that combine external knowledge sources with generative models. It highlights common methodological pipelines—input representation, core modules (retrievers, rerankers, planners, memory, tool interfaces), learning strategies (supervised fine-tuning, contrastive learning, distillation, reinforcement learning), and inference strategies such as iterative retrieval and budget control. The discussion also examines the relationship between SIGIR and related venues like RecSys, the shift from static nDCG metrics to task success and citation accuracy, agentic search paradigms, and open problems in evaluation reliability, latency, cost, hallucination, and multilingual or multimodal extension. Cross-references to related workshops on eCommerce, multimodal search, and generative AI personalization are included, making this a useful hub for researchers and engineers tracking IR trends in 2025.

SIGIR 2025

Official site: https://sigir2025.dei.unipd.it/

SIGIR 2025 is the flagship international conference on research and development in information retrieval, covering search, recommendation, and personalization in the LLM era.

Key points

Focus areas: open-domain information access, enterprise knowledge retrieval, conversational search, semantic understanding in recommendation systems, and end-to-end architectures combining external knowledge sources with generative models.
Typical methodological pipeline:

Input & representation

Core modules

Learning strategies

Inference strategies

Insights for search, recommendation, and personalization

1. Architecture: cascaded retrieval + rerank + generation remains mainstream, but agentic paradigms are making retrieval count and policy itself a learnable object. 2. Data: high-quality instruction data and click/session logs are both critical; synthetic data requires safeguards against knowledge leakage and distribution shift. 3. Evaluation: the gap between offline metrics and online satisfaction is widening; LLM-as-judge must be cross-validated with human assessment. 4. Product: latency, cost, explainability, and safety are hard constraints for industrial deployment — academic benchmarks alone are insufficient.

Limitations and open problems

Common limitations include experiment scale constrained by GPU budgets, benchmarks misaligned with real user distributions, English-centric data limiting cross-lingual generalization, and security risks of agentic systems on the open web. Future directions include more efficient test-time compute allocation, deeper integration with knowledge graphs and structured databases, and causal/fairness constraints for recommendation systems.

Engineering checklist

| Check | Question | Recommendation | |-------|----------|----------------| | Data | PII in training/index? Version control? | Partitioned indexes, anonymization, rollback-capable embedding versions | | Latency | p99 budget? Retrieval steps? | Cascades + early stop, hot-query caching, async reranking | | Quality | Do offline gains translate to CTR/satisfaction? | Interleaving experiments, human audits, citation verification | | Safety | Poisoning/bias from open retrieval? | Source whitelists, adversarial detection, output filtering | | Cost | Token & GPU cost per query? | Small-model routing, distillation, hybrid sparse+dense retrieval |

Glossary

| Term | Meaning | |------|---------| | IR | Information Retrieval | | RAG | Retrieval-Augmented Generation | | LTR | Learning to Rank | | nDCG | Normalized Discounted Cumulative Gain | | Agentic Search | Search modeled as sequential decision-making and tool invocation | | Gen-IR | Generative Information Retrieval |

Related entries

2025 SIGIR Workshop on eCommerce
CIKM 2024 1st Workshop on Multimodal Search and Recommendations
EACL 2024 Workshop on Personalization of Generative AI Systems
ICDM MMSR 2025

References

SIGIR 2025. Official conference site: https://sigir2025.dei.unipd.it/

> Note for practitioners: treat paper conclusions as directional evidence; validate with shadow deployments and A/B tests before full rollout, accounting for query distribution drift, seasonality, cold-start items, and multilingual user mixes.