Is BadChain superseded?

BadChain (Retrieval-augmented generation): superseded — cited as a baseline and beaten by newer methods. 1 paper(s) critique it, 1 beat it on benchmarks — #116 of 1179 most-superseded. Sub-problem: cluster led by PoisonedRAG. Newer alternatives in the same sub-problem include DiscourseFlip, SilentRetrieval, Deceptive Evolutionary Jamming Attack (DEJA), LogicPoison, KEPo.

Method Drift›Retrieval-augmented generation

Superseded baseline#116 of 1,179 most-superseded

BadChain

BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models

Retrieval-augmented generation · first seen Jan 20, 2024

superseded — cited as a baseline and beaten by newer methods

1 papers critique it · 1 beat it on benchmarks

What papers say

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

“backdoor-based methods lead the augmented LLM to generate incorrect answers or decisions when watermark phrases appear, making them susceptible to anomaly detection and harmful”
— Towards Copyright Protection for Knowledge Bases of Retrieval-augmented Language Models via Reasoning

Beaten on benchmarks

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

[name]-O beats BadChain · H [NQ dataset]
0.17 vs 0.85
Towards Copyright Protection for Knowledge Bases of Retrieval-augmented Language Models via Reasoning
[name]-O beats BadChain · H [HotpotQA dataset]
0.10 vs 0.84
Towards Copyright Protection for Knowledge Bases of Retrieval-augmented Language Models via Reasoning
[name]-O beats BadChain · H [MS-MARCO dataset]
0.15 vs 0.82
Towards Copyright Protection for Knowledge Bases of Retrieval-augmented Language Models via Reasoning

Newer alternatives

Recent methods in the same sub-problem, not yet superseded in the knowledge base.