A Few Brief Notes on DeepImpact, COIL, and a Conceptual Framework for Information Retrieval Techniques
read the original abstract
Recent developments in representational learning for information retrieval can be organized in a conceptual framework that establishes two pairs of contrasts: sparse vs. dense representations and unsupervised vs. learned representations. Sparse learned representations can further be decomposed into expansion and term weighting components. This framework allows us to understand the relationship between recently proposed techniques such as DPR, ANCE, DeepCT, DeepImpact, and COIL, and furthermore, gaps revealed by our analysis point to "low hanging fruit" in terms of techniques that have yet to be explored. We present a novel technique dubbed "uniCOIL", a simple extension of COIL that achieves to our knowledge the current state-of-the-art in sparse retrieval on the popular MS MARCO passage ranking dataset. Our implementation using the Anserini IR toolkit is built on the Lucene search library and thus fully compatible with standard inverted indexes.
This paper has not been read by Pith yet.
Forward citations
Cited by 12 Pith papers
-
AgentIR: A Workload-Adaptive Cascade Retrieval Substrate for Long-Term Conversational Memory
AgentIR introduces a workload-adaptive cascade router and time-partitioned index for long-term conversational memory retrieval that skips dense retrieval on most queries while preserving accuracy and scaling latency i...
-
M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation
M3-Embedding is a single model for multi-lingual, multi-functional, and multi-granular text embeddings trained via self-knowledge distillation that achieves new state-of-the-art results on multilingual, cross-lingual,...
-
GPUSparse: GPU-Accelerated Learned Sparse Retrieval with Parallel Inverted Indices
GPUSparse delivers exact GPU-accelerated sparse retrieval with 235x CPU speedup on MS MARCO while matching accuracy.
-
A Reproducible Benchmark and Evidence-Retrieval Software Framework for Silicon Detector R&D Literature
Presents the first evidence-grounded retrieval benchmark and hybrid RAG framework for silicon pixel detector R&D, with evaluation showing hybrid sparse-dense retrieval most reliable for evidence recovery.
-
Understanding Wacky Weights: A Dissection of SPLADE's Learned Term Importance
SPLADE models produce wacky expansion terms whose prevalence rises with larger vocabularies and falls with stricter sparsity; these terms primarily aid in-domain retrieval rather than out-of-domain generalization.
-
MEMTIER: Tiered Memory Architecture and Retrieval Bottleneck Analysis for Long-Running Autonomous AI Agents
MEMTIER delivers 38% accuracy on the 500-question LongMemEval-S benchmark with a 7B model on 6GB GPU, a 33-point gain over full-context baselines, via structured episodic memory, five-signal retrieval, and semantic co...
-
Why Advanced Encoders Lag on Sparse Retrieval? The Answer and an Approach to Bridging Vocabulary Gaps
Transferring modern encoders to normalized (lowercased) vocabularies via geometric embedding initialization and activation calibration closes the performance gap in learned sparse retrieval, achieving 52.4 nDCG on BEIR.
-
A Reproducible Benchmark and Evidence-Retrieval Software Framework for Silicon Detector R&D Literature
Introduces a reproducible benchmark and hybrid sparse-dense retrieval framework for evidence-grounded access to silicon detector literature, reporting Hit@5 of 0.917 on core queries.
-
MEMTIER: Tiered Memory Architecture and Retrieval Bottleneck Analysis for Long-Running Autonomous AI Agents
MEMTIER is a tripartite memory system for autonomous agents that raises accuracy on the LongMemEval-S benchmark from 5% to 38% via episodic JSONL storage, five-signal retrieval, and PPO-adapted weights.
-
A Reproducible Benchmark and Evidence-Retrieval Software Framework for Silicon Detector R&D Literature
Hybrid sparse-dense retrieval achieves Hit@5 of 0.917 on a new curated benchmark of silicon detector papers with released code and annotations.
-
When More Documents Hurt RAG: Mitigating Vector Search Dilution with Domain-Scoped, Model-Agnostic Retrieval
Domain scoping with organizational metadata mitigates vector search dilution in RAG and improves P@10 from 0.77 to 0.86 on 200 expert-validated queries across multiple LLMs and corpora.
-
MEMTIER: Tiered Memory Architecture and Retrieval Bottleneck Analysis for Long-Running Autonomous AI Agents
MEMTIER reports 0.382 accuracy and 0.412 F1 on the 500-question LongMemEval-S benchmark, a 33pp gain over full-context baseline using tiered memory and retrieval components on 6GB GPU hardware.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.