{"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/repeval-effective-text-evaluation-with-llm","title":"RepEval: Effective Text Evaluation with LLM Representation","arxiv_id":"2404.19563","date":"2024-04-30","proceeding":null,"authors":["Shuqian Sheng","Yi Xu","Tianhang Zhang","Zanwei Shen","Luoyi Fu","Jiaxin Ding","Lei Zhou","Xiaoying Gan","Xinbing Wang","Chenghu Zhou"],"abstract":"The era of Large Language Models (LLMs) raises new demands for automatic evaluation metrics, which should be adaptable to various application scenarios while maintaining low cost and effectiveness. Traditional metrics for automatic text evaluation are often tailored to specific scenarios, while LLM-based evaluation metrics are costly, requiring fine-tuning or rely heavily on the generation capabilities of LLMs. Besides, previous LLM-based metrics ignore the fact that, within the space of LLM representations, there exist direction vectors that indicate the estimation of text quality. To this end, we introduce RepEval, a metric that leverages the projection of LLM representations for evaluation. Through simple prompt modifications, RepEval can easily transition to various tasks, requiring only minimal sample pairs for direction vector construction. Results on fourteen datasets across two evaluation tasks demonstrate the high effectiveness of our method, which exhibits a higher correlation with human judgments than previous methods, even in complex evaluation scenarios involving pair-wise selection under nuanced aspects. Our work underscores the richness of information regarding text quality embedded within LLM representations, offering insights for the development of new metrics.","url_abs":"https://arxiv.org/abs/2404.19563v2","url_pdf":"https://arxiv.org/pdf/2404.19563v2.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":"repeval-effective-text-evaluation-with-llm","repo_url":"https://github.com/susisheng/repeval","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"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":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"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":"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"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2404.19563","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2404.19563"}},"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":"deterministic:regex_extraction","url":"https://github.com/neulab/BARTScore","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/shikib/usr","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/Tiiiger/bert_score","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/jinlanfu/GPTScore","reach":{"status":"ok"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/maszhongming/UniEval","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":3,"ran":13,"ran_honours":2,"unverified":9},"by_repo_kind":{"found_in_text":{"samples":27,"ran":18,"repositories":4}},"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":3,"samples":[{"code_sha256_prefix":"0f786c407fb1ee4c","entry":"swish","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_openai.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_openai.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0f786c407fb1ee4c"}},{"code_sha256_prefix":"1a5cb15d63f16098","entry":"add_question","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/utils.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":"1a5cb15d63f16098"}},{"code_sha256_prefix":"671a619bed5669a3","entry":"angle_defn","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_ctrl.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_ctrl.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"671a619bed5669a3"}},{"code_sha256_prefix":"31b107371370f1bf","entry":"calculate_correlation","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"reproduce/correlation.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/reproduce/correlation.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":"31b107371370f1bf"}},{"code_sha256_prefix":"40a6d254f3b50b99","entry":"convert_to_json","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/utils.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":"40a6d254f3b50b99"}},{"code_sha256_prefix":"8d23fbe2b99b840b","entry":"gelu","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_gpt2.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_gpt2.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8d23fbe2b99b840b"}},{"code_sha256_prefix":"211753ba29188d4e","entry":"gelu","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_bert.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_bert.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"211753ba29188d4e"}},{"code_sha256_prefix":"69c6d14de8190cfb","entry":"gelu","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_distilbert.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_distilbert.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"69c6d14de8190cfb"}},{"code_sha256_prefix":"77601724cad03f95","entry":"gelu_new","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_bert.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_bert.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"77601724cad03f95"}},{"code_sha256_prefix":"729c770d517b1bc6","entry":"get_unique_value","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"reproduce/correlation.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/reproduce/correlation.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":"729c770d517b1bc6"}},{"code_sha256_prefix":"cdc21dc661030083","entry":"load_data","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"pseudo_data_summ.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/pseudo_data_summ.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":"cdc21dc661030083"}},{"code_sha256_prefix":"79ec7dfe27ceef16","entry":"load_json","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"reproduce/data_utils.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/reproduce/data_utils.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":"79ec7dfe27ceef16"}},{"code_sha256_prefix":"154c40d3e5f52366","entry":"load_output","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"billboard/evaluate.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/billboard/evaluate.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":"154c40d3e5f52366"}},{"code_sha256_prefix":"7f6c2b0cc66bbc7c","entry":"load_src","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"billboard/evaluate.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/billboard/evaluate.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":"7f6c2b0cc66bbc7c"}},{"code_sha256_prefix":"0c145728a35789fa","entry":"load_tf_weights_in_bert","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_bert.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_bert.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0c145728a35789fa"}},{"code_sha256_prefix":"59ac77747a12b1cb","entry":"positional_encoding","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_ctrl.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_ctrl.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"59ac77747a12b1cb"}},{"code_sha256_prefix":"57ebc1447d921c46","entry":"scaled_dot_product_attention","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_ctrl.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_ctrl.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"57ebc1447d921c46"}},{"code_sha256_prefix":"a878dc3112f9ce9e","entry":"tile","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_beam_search.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_beam_search.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a878dc3112f9ce9e"}},{"code_sha256_prefix":"3186cad3f4fa28a4","entry":"BARTScorer","repo":"neulab/BARTScore","repo_kind":"found_in_text","path":"bart_score.py","file_url":"https://github.com/neulab/BARTScore/blob/HEAD/bart_score.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3186cad3f4fa28a4"}},{"code_sha256_prefix":"388a367eded0b858","entry":"GPT3Model","repo":"jinlanfu/GPTScore","repo_kind":"found_in_text","path":"score_d2t.py","file_url":"https://github.com/jinlanfu/GPTScore/blob/HEAD/score_d2t.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"388a367eded0b858"}},{"code_sha256_prefix":"70093442b3ee191f","entry":"Scorer","repo":"jinlanfu/GPTScore","repo_kind":"found_in_text","path":"score_d2t.py","file_url":"https://github.com/jinlanfu/GPTScore/blob/HEAD/score_d2t.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"70093442b3ee191f"}},{"code_sha256_prefix":"666cfad4555e9faa","entry":"disfluency_transformation","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"pseudo_data_summ.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/pseudo_data_summ.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":"666cfad4555e9faa"}},{"code_sha256_prefix":"6b02658358cb6a62","entry":"gpt3score","repo":"jinlanfu/GPTScore","repo_kind":"found_in_text","path":"score_d2t.py","file_url":"https://github.com/jinlanfu/GPTScore/blob/HEAD/score_d2t.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6b02658358cb6a62"}},{"code_sha256_prefix":"bb6e70aa3a9fcf4e","entry":"incoherence_transformation","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"pseudo_data_summ.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/pseudo_data_summ.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":"bb6e70aa3a9fcf4e"}},{"code_sha256_prefix":"dbd9587f7a986fc3","entry":"load_ref","repo":"maszhongming/UniEval","repo_kind":"found_in_text","path":"billboard/evaluate.py","file_url":"https://github.com/maszhongming/UniEval/blob/HEAD/billboard/evaluate.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":"dbd9587f7a986fc3"}},{"code_sha256_prefix":"654e91efc679fe84","entry":"load_tf_weights_in_gpt2","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_gpt2.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_gpt2.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"654e91efc679fe84"}},{"code_sha256_prefix":"42ee168c3ae12e60","entry":"load_tf_weights_in_openai_gpt","repo":"shikib/usr","repo_kind":"found_in_text","path":"transformers/modeling_openai.py","file_url":"https://github.com/shikib/usr/blob/HEAD/transformers/modeling_openai.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"42ee168c3ae12e60"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}