# thunlp/TAADpapers

Must-read Papers on Textual Adversarial Attack and Defense

Repository: https://github.com/thunlp/TAADpapers
Canonical: https://ross.abutalabs.com/products/taadpapers
Language: Python
License: MIT
License Family: permissive
Topics: paper-list, nlp, adversarial-learning, adversarial-attacks, adversarial-defense, natural-language-processing
Last push: 2025-06-04T15:03:06+00:00

## Health v2 (maintenance only)
Score: 44/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 25, release rhythm 35, longevity 100
- inputs: {"age_days": 2642, "days_push": 455, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1575, forks 194 (observed 2026-08-28T04:05:05.974329+00:00)

## What it is
A curated, categorized list of must-read research papers on textual adversarial attack and defense in NLP. It organizes over 150 papers by perturbation level, defense, certified robustness, benchmarks, and toolkits.

## Use cases
- find papers on textual adversarial attacks in NLP
- survey adversarial defense methods for language models
- research robustness of NLP models
- find toolkits for generating adversarial text examples
- get started with adversarial NLP research
- find benchmarks for evaluating adversarial robustness in text

## When to choose
- you need a comprehensive, categorized reading list on textual adversarial attack and defense
- you are starting research on NLP model robustness
- you want to find related toolkits like OpenAttack or TextAttack

## When to avoid
- you need runnable software rather than a paper list
- you need coverage of adversarial attacks outside the text/NLP domain
- you need a complete bibliography with full abstracts or PDFs hosted locally

## Facets
- artifact type: learning-resource
- maturity: active
- function: nlp, security, machine-learning
- domain: security, machine-learning, awesome-lists
- platform: cross-platform
- tags: paper-list, adversarial-attacks, adversarial-defense, robustness, curated-list, natural-language-processing

## Member repositories
- thunlp/TAADpapers (main) score 44

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:05.974329+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:57:36.739324+00:00, confidence not recorded.
  - readme: https://github.com/thunlp/TAADpapers (fetched 2026-08-28T04:05:05.974329+00:00, sha 7334c592e281)
- Data as of 2026-08-30T08:39:29.467469+00:00.