{"about":{"site":"https://codewithpapers.app","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.","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"},"url":"/paper/wait-that-s-not-an-option-llms-robustness","title":"Wait, that's not an option: LLMs Robustness with Incorrect Multiple-Choice Options","arxiv_id":"2409.00113","date":"2024-08-27","proceeding":null,"authors":["Gracjan Góral","Emilia Wiśnios","Piotr Sankowski","Paweł Budzianowski"],"abstract":"Decision-making under full alignment requires balancing between reasoning and faithfulness - a challenge for large language models (LLMs). This study explores whether LLMs prioritize following instructions over reasoning and truth when given \"misleading\" instructions, such as \"Respond solely with A or B\", even when neither option is correct. We introduce a new metric called \"reflective judgment\", which sheds new light on the relationship between the pre-training and post-training alignment schemes. In tasks ranging from basic arithmetic to domain-specific assessments, models like GPT-4o, o1-mini, or Claude 3 Opus adhered to instructions correctly but failed to reflect on the validity of the provided options. Contrary, models from the Llama 3.1 family (8B, 70B, 405B) or base Qwen2.5 (7B, 14B, 32B) families exhibit improved refusal rates with size, indicating a scaling effect. We also observed that alignment techniques, though intended to enhance reasoning, sometimes weakened the models' ability to reject incorrect instructions, leading them to follow flawed prompts uncritically. Finally, we have also conducted a parallel human study revealing similar patterns in human behavior and annotations. We highlight how popular RLHF datasets might disrupt either training or evaluation due to annotations exhibiting poor reflective judgement.","url_abs":"https://arxiv.org/abs/2409.00113v2","url_pdf":"https://arxiv.org/pdf/2409.00113v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"wait-that-s-not-an-option-llms-robustness","repo_url":"https://github.com/gracjangoral/when-all-options-are-wrong","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"multiple-choice","task_name":"Multiple-choice"}],"methods":[{"method_slug":"base","method_name":"BASE"},{"method_slug":"llama","method_name":"LLaMA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2409.00113","atlas_url":"https://app.syntology.ai/?focus=2409.00113","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2409.00113"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gracjangoral/when-all-options-are-wrong","reach":{"status":"ok","spdx":"GPL-3.0"}}],"summary":{"ran":2},"by_repo_kind":{"official":{"samples":2,"ran":2,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":2,"samples":[{"code_sha256_prefix":"730f8a40beb44480","entry":"generate_dataset","repo":"gracjangoral/when-all-options-are-wrong","repo_kind":"official","path":"math_dataset_generation.py","file_url":"https://github.com/gracjangoral/when-all-options-are-wrong/blob/HEAD/math_dataset_generation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"730f8a40beb44480"}},{"code_sha256_prefix":"4b4b7e196e0334b3","entry":"generate_example","repo":"gracjangoral/when-all-options-are-wrong","repo_kind":"official","path":"math_dataset_generation.py","file_url":"https://github.com/gracjangoral/when-all-options-are-wrong/blob/HEAD/math_dataset_generation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"4b4b7e196e0334b3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}