{"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/imageinwords-unlocking-hyper-detailed-image","title":"ImageInWords: Unlocking Hyper-Detailed Image Descriptions","arxiv_id":"2405.02793","date":"2024-05-05","proceeding":null,"authors":["Roopal Garg","Andrea Burns","Burcu Karagol Ayan","Yonatan Bitton","Ceslee Montgomery","Yasumasa Onoe","Andrew Bunner","Ranjay Krishna","Jason Baldridge","Radu Soricut"],"abstract":"Despite the longstanding adage \"an image is worth a thousand words,\" generating accurate hyper-detailed image descriptions remains unsolved. Trained on short web-scraped image text, vision-language models often generate incomplete descriptions with visual inconsistencies. We address this via a novel data-centric approach with ImageInWords (IIW), a carefully designed human-in-the-loop framework for curating hyper-detailed image descriptions. Human evaluations on IIW data show major gains compared to recent datasets (+66%) and GPT4V (+48%) across comprehensiveness, specificity, hallucinations, and more. We also show that fine-tuning with IIW data improves these metrics by +31% against models trained with prior work, even with only 9k samples. Lastly, we evaluate IIW models with text-to-image generation and vision-language reasoning tasks. Our generated descriptions result in the highest fidelity images, and boost compositional reasoning by up to 6% on ARO, SVO-Probes, and Winoground datasets. We release the IIW Eval benchmark with human judgement labels, object and image-level annotations from our framework, and existing image caption datasets enriched via IIW-model.","url_abs":"https://arxiv.org/abs/2405.02793v2","url_pdf":"https://arxiv.org/pdf/2405.02793v2.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":"imageinwords-unlocking-hyper-detailed-image","repo_url":"https://github.com/google/imageinwords","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"imageinwords-unlocking-hyper-detailed-image","repo_url":"https://github.com/google-deepmind/proactive_t2i_agents","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"specificity","task_name":"Specificity"},{"task_slug":"text-to-image-generation-1","task_name":"Text to Image Generation"},{"task_slug":"text-to-image-generation","task_name":"Text-to-Image Generation"}],"methods":[],"datasets_introduced":[{"slug":"iiw-400","name":"IIW-400","full_name":"ImageInWords: IIW-400"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2405.02793","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.02793"}},"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/google/imageinwords","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-deepmind/proactive_t2i_agents","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":10},"by_repo_kind":{"listed":{"samples":10,"ran":0,"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":"6fdfee8ae85731ff","entry":"entity_parser_prompt","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/prompts.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/prompts.py","link_basis":"harvester_set","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":"6fdfee8ae85731ff"}},{"code_sha256_prefix":"96bd654b5fe8cafd","entry":"entity_parser_prompt_w_background","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/prompts.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/prompts.py","link_basis":"harvester_set","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":"96bd654b5fe8cafd"}},{"code_sha256_prefix":"ac24c3867c0d6c29","entry":"entropy","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/agent_classes.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/agent_classes.py","link_basis":"harvester_set","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":"ac24c3867c0d6c29"}},{"code_sha256_prefix":"7d93f6d7eb593350","entry":"get_image_description","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/image_utils.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/image_utils.py","link_basis":"harvester_set","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":"7d93f6d7eb593350"}},{"code_sha256_prefix":"4337cf11cd0ec49c","entry":"is_confident","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/belief_utils.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/belief_utils.py","link_basis":"harvester_set","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":"4337cf11cd0ec49c"}},{"code_sha256_prefix":"912864da44ac4f8d","entry":"load_img","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/image_utils.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/image_utils.py","link_basis":"harvester_set","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":"912864da44ac4f8d"}},{"code_sha256_prefix":"9969f5aeb6be4c91","entry":"merge_info_prompt","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/prompts.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/prompts.py","link_basis":"harvester_set","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":"9969f5aeb6be4c91"}},{"code_sha256_prefix":"2d562b2e2b98bf22","entry":"most_likely_deterministic_state","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/belief_utils.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/belief_utils.py","link_basis":"harvester_set","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":"2d562b2e2b98bf22"}},{"code_sha256_prefix":"b9dcd4f5f46998b9","entry":"negative_log_likelihood","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/belief_metrics.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/belief_metrics.py","link_basis":"harvester_set","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":"b9dcd4f5f46998b9"}},{"code_sha256_prefix":"dce5b9144b3351bc","entry":"show_image","repo":"google-deepmind/proactive_t2i_agents","repo_kind":"listed","path":"agent/image_utils.py","file_url":"https://github.com/google-deepmind/proactive_t2i_agents/blob/HEAD/agent/image_utils.py","link_basis":"harvester_set","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":"dce5b9144b3351bc"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}