Is LongLLMLingua superseded?

LongLLMLingua (Retrieval-augmented generation): superseded — cited as a baseline and beaten by newer methods. 3 paper(s) critique it, 12 beat it on benchmarks — #18 of 1179 most-superseded.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Superseded baseline#18 of 1,179 most-superseded

LongLLMLingua

LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression

Retrieval-augmented generation · first seen Oct 10, 2023

superseded — cited as a baseline and beaten by newer methods

3 papers critique it · 12 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites LongLLMLingua as a baseline.

“While some methods, such as perplexity-based approaches, can process longer texts, they often lack comprehensive document understanding.”
— Hierarchical Document Refinement for Long-context Retrieval-augmented Generation
“While these methods improve textual RAG, they do not address the fundamental cross-modal attention collapse that occurs in multimodal architectures.”
— The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation
“the pre-processing methods introduce additional computational costs during inference and may lead to the loss of essential information.”
— R^2AG: Incorporating Retrieval Information into Retrieval Augmented Generation

Beaten on benchmarks

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

LongRefiner beats LongLLMLingua · Accuracy [2k token limit, Qwen2.5-7B-Instruct]
50.6 vs 40.4
Hierarchical Document Refinement for Long-context Retrieval-augmented Generation
LongRefiner beats LongLLMLingua · F1 [2k token limit, Qwen2.5-7B-Instruct]
40.6 vs 33.9
Hierarchical Document Refinement for Long-context Retrieval-augmented Generation
LongLLMLingua + BAIR beats LongLLMLingua · Acc. [MedGemma-4B (IU-Chest)]
65.91 vs 65.66
The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation
LongLLMLingua + BAIR beats LongLLMLingua · Acc. [CheXagent-8B (IU-Chest)]
66.24 vs 65.33
The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation
LongLLMLingua + BAIR beats LongLLMLingua · Acc. [Qwen2.5-VL-3B (FACET)]
87.45 vs 84.36
The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation
LongLLMLingua + BAIR beats LongLLMLingua · Acc. [DeepSeek-VL-7B (FACET)]
87.80 vs 87.76
The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation
LongLLMLingua + BAIR beats LongLLMLingua · Acc. [SkySenseGPT-7B (NWPU)]
70.79 vs 69.97
The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation
LongLLMLingua + BAIR beats LongLLMLingua · Acc. [EarthDial (NWPU)]
94.49 vs 94.38
The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation
AttnComp beats LongLLMLingua · Acc [HotpotQA]
45.2 vs 39.1
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats LongLLMLingua · Acc [2WikiMQA]
38.1 vs 31.9
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats LongLLMLingua · Acc [MuSiQue]
19.6 vs 13.8
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
AttnComp beats LongLLMLingua · Acc [NQ]
53.0 vs 48.1
AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation

Newer alternatives

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