Is CompAct superseded?

CompAct (Retrieval-augmented generation): superseded — cited as a baseline and beaten by newer methods. 0 paper(s) critique it, 5 beat it on benchmarks — #37 of 1179 most-superseded. Sub-problem: cluster led by RECOMP. Newer alternatives in the same sub-problem include Bottleneck Attention Intervention for Recovery (BAIR), QREAM, CoCR-RAG, XProvence, Crucible.

Method Drift›Retrieval-augmented generation

Superseded baseline#37 of 1,179 most-superseded

CompAct

CompAct: Compressing Retrieved Documents Actively for Question Answering

Retrieval-augmented generation · first seen Jul 12, 2024

superseded — cited as a baseline and beaten by newer methods

0 papers critique it · 5 beat it on benchmarks

Beaten on benchmarks

Head-to-head results where a newer method reports beating CompAct. Values are copied from the source paper's tables — verify against the cited paper.

Oreo beats CompAct · EM [TriviaQA single-hop]
0.4257 vs 0.4242
Oreo: A Plug-in Context Reconstructor to Enhance Retrieval-Augmented Generation
Oreo beats CompAct · F1 [Multi-hop QA with Flan-T5]
0.658 vs 0.5974
Oreo: A Plug-in Context Reconstructor to Enhance Retrieval-Augmented Generation
AttnComp beats CompAct · Acc [HotpotQA]
45.2 vs 40.2
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats CompAct · Acc [2WikiMQA]
38.1 vs 33.2
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats CompAct · Acc [MuSiQue]
19.6 vs 16.5
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats CompAct · Acc [NQ]
53.0 vs 48.7
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats CompAct · Acc [PopQA]
65.1 vs 58.4
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats CompAct · Acc [AVG]
44.2 vs 39.4
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
QREAM-FT beats CompAct · Accuracy [Standard RAG Pipeline with Llama-3-8B-Instruct]
45.6 vs 40.1
Align Documents to Questions: Question-Oriented Document Rewriting for Retrieval-Augmented Generation
QREAM-FT beats CompAct · Accuracy [Standard RAG Pipeline with Mistral-7B-Instruct]
44.9 vs 39.8
Align Documents to Questions: Question-Oriented Document Rewriting for Retrieval-Augmented Generation
Ext2Gen-R2 beats CompAct · Acc [Llama3.1-8b-instruct backbone]
0.463 vs 0.343
Aligning Extraction and Generation for Robust Retrieval-Augmented Generation
Ext2Gen-R2 beats CompAct · LLMEval [Llama3.1-8b-instruct backbone]
0.860 vs 0.736
Aligning Extraction and Generation for Robust Retrieval-Augmented Generation

Newer alternatives

Recent methods in the same sub-problem, not yet superseded in the knowledge base.