Entity · technique
BM25
techniqueactive
bm25-bde08fee·8 events·first seen Aliases: BM25
Co-occurring entities
Model Context ProtocolAPS-RAGAdvanced Photon SourceAPS-BenchU.S. Nuclear Regulatory CommissionGoogleRetrieval-Augmented Fine-TuningGemma 4Retrieval-Augmented GenerationNRC Generic Fundamentals ExaminationHotpotQAObservable Entity RelevanceCounterfactual Trajectory UtilityBridge Evidence: Static Retrieval Utility Does Not Predict Causal Utility in Multi-Step Agentic SearchPubMedBGEReciprocal Rank FusionEurope PMCBioASQ 14bGPT-5.5
More like this (12)
Recent events (8)
APS-RAG: Corrective agentic hybrid RAG system deployed at Advanced Photon Source with operations-grounded evaluation
Researchers present APS-RAG, a deployed retrieval-augmented generation platform for the Advanced Photon Source (APS) scientific facility that fuses dense, sparse, and knowledge-graph retrieval channels with a corrective agentic loop and a ReAct executor over an MCP tooling layer. The system is evaluated on APS-Bench, a new 50-question QA dataset with auditable gold answers, where the full corrective Agentic GraphRAG achieves 70.3% strict vital-nugget recall versus 63.8% for a BM25 baseline. The cross-encoder reranker is identified as the most impactful component, with its removal causing a 32.8% drop in recall. The codebase, evaluation harness, and benchmark construction methodology are released to support adoption at other scientific facilities.
Gemma 4 31B benchmarked on NRC Reactor Operator Licensing Exam via fine-tuning and RAG configurations
Researchers benchmark eight fine-tuning and retrieval configurations of Gemma 4 31B-IT against the U.S. NRC Reactor Operator licensing examination across 14 Generic Fundamentals Examinations from 2015–2021. The best configuration—supervised fine-tuning with Gemini-distilled chain-of-thought rationales combined with fixed-size chunking RAG—passed 8 of 14 exams and reached 79.7% aggregate accuracy, near the 80% human passing threshold. Key findings include that no configuration without fine-tuning passed any exam, that preferred chunking strategy reverses depending on training state, and that RAFT underperforms standard SFT in matched retrieval environments.
Bridge Evidence: Static Retrieval Utility Does Not Predict Causal Utility in Multi-Step Agentic Search
A new arXiv paper demonstrates empirically that static retrieval relevance scores are nearly statistically independent from causal utility in multi-step agentic search (Spearman rho = -0.026 across 23,322 document observations). Using a ReAct-style agent over HotpotQA with counterfactual trajectory replays, the authors show that roughly a third of documents are causally load-bearing ('bridge documents') while appearing useless to a static reader. The mechanism is identified: bridge documents supply discriminative entities that redirect subsequent queries, meaning optimizing static RAG utility does not deliver agentic utility.
Cost-Pragmatic Retrieval Gating and Multi-Model Fusion for BioASQ 2026 Biomedical QA
A BioASQ Task 14B 2026 system paper describes two core design decisions: a cost-pragmatic re-retrieval policy using a BGE cross-encoder quality gate, and a decomposition of multi-model ensemble lift into selection and fusion components. The retrieval pipeline unions dense (BGE + BM25 + RRF) and agent-driven PubMed/Europe PMC/iCite pipelines, achieving R@200 = 99.3% on the BioASQ-13b archive. The team places first on the combined-exact aggregate on three of eight leaderboard tracks and first on Phase B b3 ideal. GPT-5.5 solo retains list-F1 lead over a synonym-union resolver due to precision trade-offs.
Tool-intent stabilization analysis quantifies when streaming RAG latency hiding is possible
A new arXiv paper introduces 'tool-intent stabilization' — the point in a streaming input at which a speculative retrieval query converges to the correct result — and measures its distribution on the CRAG benchmark (1,371 questions). The authors derive a model-agnostic bound on how much tool latency can be hidden behind remaining user input, finding that at realistic operating parameters 73.9% of queries admit substantial latency hiding. The study requires no model training and validates the bound against a working streaming pipeline, also identifying query properties that predict early versus late stabilization.
Anthropic introduces Contextual Retrieval to reduce RAG retrieval failures by up to 67%
Anthropic published a technical method called Contextual Retrieval that combines Contextual Embeddings and Contextual BM25 to address the context-loss problem in traditional RAG pipelines. The approach prepends chunk-level context before encoding, reducing failed retrievals by 49% standalone and 67% when combined with reranking. The post also highlights prompt caching as a simpler alternative for knowledge bases under 200K tokens, and provides a cookbook for deployment with Claude.
Mistral Releases Search Toolkit: Open-Source Composable Framework for Production RAG and Enterprise Search Pipelines
Mistral AI has launched Search Toolkit in public preview, an open-source framework that unifies document ingestion, retrieval, and evaluation into a single composable pipeline for AI applications. The toolkit ships with BM25 sparse retrieval, dense embedding-based retrieval, hybrid configurations, and built-in metrics (recall, precision, MRR, NDCG), targeting enterprise RAG workflows, domain-specific retrieval, and agentic systems. It integrates with MCP-based Connectors for live data access from CRMs, code repositories, and productivity tools. CMA CGM is cited as a production user, combining Search Toolkit with Voxtral for real-time fake news detection across audio sources.
SPECTRA: Synthetic IR Test Collections with Relevance Oracles and Controlled Distractor Diagnostics
SPECTRA is a reproducible framework for generating synthetic information retrieval test collections, separating latent topical structure, surface text realization, and query intent generation to produce deterministic relevance oracles without human annotation. A Python prototype generated corpora up to 60,000 documents at roughly 12K–14K documents per second, with graded relevance labels for 96 queries. Controlled distractor experiments showed BM25 nDCG@10 degrading from 1.00 at 2% distractors to 0.43 at 36%, demonstrating the framework's utility for exposing retrieval system failure modes before expensive real-world collection construction. The authors position SPECTRA as a diagnostic complement to Cranfield/TREC-style evaluation rather than a replacement for human judgment.