{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/dataset/redeval","name":"RedEval","full_name":null,"description_markdown":"**RedEval** is a safety evaluation benchmark designed to assess the robustness of large language models (LLMs) against harmful prompts. It simulates and evaluates LLM applications across various scenarios, all while **eliminating the need for human intervention**. Here are the key aspects of RedEval:\r\n\r\n1. **Purpose**: RedEval aims to evaluate LLM safety using a technique called **Chain of Utterances (CoU)**-based prompts. CoU prompts are effective at breaking the safety guardrails of various LLMs, including **GPT-4**, **ChatGPT**, and open-source models.\r\n\r\n2. **Safety Assessment**: RedEval provides **simple scripts** to evaluate both **closed-source systems** (such as ChatGPT and GPT-4) and **open-source LLMs** on its benchmark. The evaluation focuses on **harmful questions** and computes the **Attack Success Rate (ASR)**.\r\n\r\n3. **Question Banks**:\r\n    - **HarmfulQA**: Consists of **1,960 harmful questions** covering **10 topics** and approximately **10 subtopics** each.\r\n    - **DangerousQA**: Contains **200 harmful questions** across **6 adjectives**: racist, stereotypical, sexist, illegal, toxic, and harmful.\r\n    - **CategoricalQA**: Includes **11 categories of harm**, each with **5 sub-categories**, available in English, Chinese, and Vietnamese.\r\n    - **AdversarialQA**: Provides a set of **500 instructions** to tease out harmful behaviors from the model.\r\n\r\n4. **Safety Alignment**: RedEval also offers code to perform **safety alignment** of LLMs. For instance, it aligns **Vicuna-7B** on **HarmfulQA**, resulting in a safer version of Vicuna that is more robust against RedEval.\r\n\r\n5. **Installation**:\r\n    - Create a conda environment: `conda create --name redeval -c conda-forge python=3.11`\r\n    - Activate the environment: `conda activate redeval`\r\n    - Install required packages: `pip install -r requirements.txt`\r\n    - Store API keys in the `api_keys` directory for use by the LLM as a judge and the `generate_responses.py` script for closed-source models.\r\n\r\n6. **Prompt Templates**:\r\n    - Choose a prompt template for red-teaming:\r\n        - **Chain of Utterances (CoU)**: Effective at breaking safety guardrails.\r\n        - **Chain of Thoughts (CoT)**\r\n        - **Standard prompt**\r\n        - **Suffix prompt**\r\n        - Note: Different LLMs may require slight variations in the prompt template.\r\n\r\n7. **How to Perform Red-Teaming**:\r\n    - **Step 0**: Decide on the prompt template.\r\n    - **Step 1**: Generate model outputs on harmful questions by providing a path to the question bank and the red-teaming prompt.","description_withheld":null,"homepage":"https://github.com/chziakas/RedEval","introduced_date":"2023-08-18","introduced_date_note":null,"introduced_by":{"paper":"/paper/red-teaming-large-language-models-using-chain","title":"Red-Teaming Large Language Models using Chain of Utterances for Safety-Alignment","first_author":"Rishabh Bhardwaj","url":null},"license":null,"modalities":[],"tasks":[],"languages":[],"variants":["RedEval"],"data_loaders":[],"num_papers_in_archive":2,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[],"papers":{"order":"date (newest first), then slug","population":"every paper with a leaderboard row on this dataset's benchmarks (the benchmark-backed subset): the archive's own papers-using-this-dataset list was never published, so this is not that list; num_papers_in_archive is the archive's own count","page":1,"pages_in_order":1,"rows_per_page":100,"shown":0,"of":0,"counts":{"papers_with_a_benchmark_row":0,"with_a_code_link":0,"where_syntology_ran_a_sample":0,"not_listed_spam_title":0,"listed":0,"listed_where_code_ran":0,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":0,"every_run_a_failure_of_syntologys_instrument":0,"listed_with_a_run_with_no_instrument_failure":0,"listed_every_run_a_failure_of_syntologys_instrument":0,"filter":{"states":["a run with no instrument failure","any run, instrument failures included"],"default":"a run with no instrument failure","note":"on the 'only where code ran' pages the default hides, in the browser, the rows where every run was a failure of Syntology's instrument; the second state shows them again. Rows are hidden, never re-ordered; these twins list every row"}},"definition":"distinct papers with at least one leaderboard row on this dataset's benchmarks; 'where Syntology ran a sample' counts papers with at least one harvested sample that ran, which is not a correctness claim"},"items":[],"full_list":"papers_with_a_benchmark_row: every listed paper, in this order"},"papers_with_a_benchmark_row":[],"papers_where_code_ran":0,"papers_where_code_ran_first_page":null,"papers_where_code_ran_caption":"We ran code from the paper's repository; we did not run it on this dataset or check it against this dataset's benchmarks.","pages":[],"row_route":"each paper's syntology object carries sample_list (this site's paper page, #ran), syntology_url (syntology.ai's page for the paper, only when that route probed live) and mcp (the agent call that lists the paper's harvested samples; not a verdict on the paper)","developers":"https://syntology.ai/developers","syntology_totals":{"read_at":"2026-10-05T00:36:11+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"papers_with_a_sample_that_ran":0,"papers_with_a_run_with_no_instrument_failure":0,"papers_where_every_run_was_an_instrument_failure":0,"note":"the per-paper counts of every listed paper (all pages), summed; not a rate"},"key_notes":{"samples_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","samples_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"since_archive_judged":{"papers":0,"papers_placed":0,"judged_by":{"model":"a model","models_as_recorded":[],"from":"lb_rows_run_verified.jsonl: each record's `model` field (the extractor's call that wrote decisions[])"},"note":"a model's judgement from the paper's own tables that they report on this leaderboard, not a result: no value from a paper that was not placed is printed or carried here; papers counts each paper once over these leaderboards","backfill":"8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read)","backfill_complete":false,"leaderboards":[]},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks.","record_sha256":"bd4aa338c5ff0559b7dd69299332f8ddc1cf15db9a7a2a5cea203a1c1452eb5e","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}