{"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/mt-eval-a-multi-turn-capabilities-evaluation","title":"MT-Eval: A Multi-Turn Capabilities Evaluation Benchmark for Large Language Models","arxiv_id":"2401.16745","date":"2024-01-30","proceeding":null,"authors":["Wai-Chung Kwan","Xingshan Zeng","Yuxin Jiang","YuFei Wang","Liangyou Li","Lifeng Shang","Xin Jiang","Qun Liu","Kam-Fai Wong"],"abstract":"Large language models (LLMs) are increasingly relied upon for complex multi-turn conversations across diverse real-world applications. However, existing benchmarks predominantly focus on single-turn evaluations, overlooking the models' capabilities in multi-turn interactions. To address this gap, we introduce MT-Eval, a comprehensive benchmark designed to evaluate multi-turn conversational abilities. By analyzing human-LLM conversations, we categorize interaction patterns into four types: recollection, expansion, refinement, and follow-up. We construct multi-turn queries for each category either by augmenting existing datasets or by creating new examples with GPT-4 to avoid data leakage. To study the factors impacting multi-turn abilities, we create single-turn versions of the 1170 multi-turn queries and compare performance. Our evaluation of 11 well-known LLMs shows that while closed-source models generally surpass open-source ones, certain open-source models exceed GPT-3.5-Turbo in specific tasks. We observe significant performance degradation in multi-turn settings compared to single-turn settings in most models, which is not correlated with the models' fundamental capabilities. Moreover, we identify the distance to relevant content and susceptibility to error propagation as the key factors influencing multi-turn performance. MT-Eval is released publicly to encourage future research towards more robust conversational models.","url_abs":"https://arxiv.org/abs/2401.16745v1","url_pdf":"https://arxiv.org/pdf/2401.16745v1.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":"mt-eval-a-multi-turn-capabilities-evaluation","repo_url":"https://github.com/kwanwaichung/mt-eval","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"focus","method_name":"Focus"},{"method_slug":"gpt-3","method_name":"GPT-3"},{"method_slug":"gpt-4","method_name":"GPT-4"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2401.16745","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2401.16745"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/kwanwaichung/mt-eval","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":9,"unverified":2},"by_repo_kind":{"official":{"samples":11,"ran":9,"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":0,"samples":[{"code_sha256_prefix":"2dd8e40379d7bbb9","entry":"compute_bleu","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/bleu.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/bleu.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2dd8e40379d7bbb9"}},{"code_sha256_prefix":"cdcc85ca09b00f7e","entry":"count_words","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/global_inst.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/global_inst.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cdcc85ca09b00f7e"}},{"code_sha256_prefix":"506bf38508c1c41e","entry":"get_logger","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/misc.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/misc.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"506bf38508c1c41e"}},{"code_sha256_prefix":"5da949adc04e26b8","entry":"highlight_max_scores_in_latex_table","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"calculate_score.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/calculate_score.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5da949adc04e26b8"}},{"code_sha256_prefix":"1b728b3dde890232","entry":"llama_generate","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"inference.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/inference.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1b728b3dde890232"}},{"code_sha256_prefix":"956dc4a97e4648e4","entry":"parse_construct_qa_output","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/parse.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/parse.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"956dc4a97e4648e4"}},{"code_sha256_prefix":"7f9889b5c320e542","entry":"parse_construct_relation_output","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/parse.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/parse.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7f9889b5c320e542"}},{"code_sha256_prefix":"236145905129b85b","entry":"parse_gpt4_evaluation","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/parse.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/parse.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"236145905129b85b"}},{"code_sha256_prefix":"4b3953284f740713","entry":"try_parse_float","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"calculate_score.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/calculate_score.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4b3953284f740713"}},{"code_sha256_prefix":"7661d3caced944b7","entry":"compute","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/bleu.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/bleu.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7661d3caced944b7"}},{"code_sha256_prefix":"cc33cc2941eec9c2","entry":"count_sentences","repo":"kwanwaichung/mt-eval","repo_kind":"official","path":"utils/global_inst.py","file_url":"https://github.com/kwanwaichung/mt-eval/blob/HEAD/utils/global_inst.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cc33cc2941eec9c2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}