{"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/the-world-of-an-octopus-how-reporting-bias","title":"The World of an Octopus: How Reporting Bias Influences a Language Model's Perception of Color","arxiv_id":"2110.08182","date":"2021-10-15","proceeding":null,"authors":["Cory Paik","Stéphane Aroca-Ouellette","Alessandro Roncone","Katharina Kann"],"abstract":"Recent work has raised concerns about the inherent limitations of text-only pretraining. In this paper, we first demonstrate that reporting bias, the tendency of people to not state the obvious, is one of the causes of this limitation, and then investigate to what extent multimodal training can mitigate this issue. To accomplish this, we 1) generate the Color Dataset (CoDa), a dataset of human-perceived color distributions for 521 common objects; 2) use CoDa to analyze and compare the color distribution found in text, the distribution captured by language models, and a human's perception of color; and 3) investigate the performance differences between text-only and multimodal models on CoDa. Our results show that the distribution of colors that a language model recovers correlates more strongly with the inaccurate distribution found in text than with the ground-truth, supporting the claim that reporting bias negatively impacts and inherently limits text-only training. We then demonstrate that multimodal models can leverage their visual training to mitigate these effects, providing a promising avenue for future research.","url_abs":"https://arxiv.org/abs/2110.08182v1","url_pdf":"https://arxiv.org/pdf/2110.08182v1.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":"the-world-of-an-octopus-how-reporting-bias","repo_url":"https://github.com/nala-cub/coda","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"jax","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[{"slug":"coda","name":"CoDa","full_name":"The Color Dataset"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2110.08182","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2110.08182"}},"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/openai/CLIP","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nala-cub/coda","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":14},"by_repo_kind":{"official":{"samples":14,"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":"d2121a771ce736f5","entry":"add_article","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/dataset/_templates.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/dataset/_templates.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":"d2121a771ce736f5"}},{"code_sha256_prefix":"1cb1a15754b73787","entry":"apply_preprocessing","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/preprocess_steps.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/preprocess_steps.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":"1cb1a15754b73787"}},{"code_sha256_prefix":"39f4ad8a949bcb18","entry":"build_update_fn","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/representations/models.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/representations/models.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":"39f4ad8a949bcb18"}},{"code_sha256_prefix":"79a31f93ab7a0cae","entry":"causal_step_loss","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/fwd_steps.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/fwd_steps.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":"79a31f93ab7a0cae"}},{"code_sha256_prefix":"49aa65398fbd515b","entry":"compute_metrics","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/metrics.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/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":"49aa65398fbd515b"}},{"code_sha256_prefix":"161a21838d44dc2d","entry":"feature_extraction_step_single_token","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/fwd_steps.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/fwd_steps.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":"161a21838d44dc2d"}},{"code_sha256_prefix":"98fbfaa5ff0b9d32","entry":"hf_auto_configure","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/config_util.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/config_util.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":"98fbfaa5ff0b9d32"}},{"code_sha256_prefix":"5872051564104ccd","entry":"initialize_models","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/representations/models.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/representations/models.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":"5872051564104ccd"}},{"code_sha256_prefix":"0e5d9a2adb715044","entry":"jensenshannon_div","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/metrics.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/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":"0e5d9a2adb715044"}},{"code_sha256_prefix":"3431e6b20c7cc562","entry":"keys_to_var_names","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/dataset/_templates.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/dataset/_templates.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":"3431e6b20c7cc562"}},{"code_sha256_prefix":"2023f174e590228c","entry":"mlm_step","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/fwd_steps.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/fwd_steps.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":"2023f174e590228c"}},{"code_sha256_prefix":"49a7caee9f2f5574","entry":"recursive_format","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/dataset/_templates.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/dataset/_templates.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":"49a7caee9f2f5574"}},{"code_sha256_prefix":"589169ba57e69b61","entry":"scale_metrics","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/result_util.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/result_util.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":"589169ba57e69b61"}},{"code_sha256_prefix":"323ea09458443106","entry":"try_get_index","repo":"nala-cub/coda","repo_kind":"official","path":"projects/coda/probing/_src/preprocess_steps.py","file_url":"https://github.com/nala-cub/coda/blob/HEAD/projects/coda/probing/_src/preprocess_steps.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":"323ea09458443106"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}