{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T06:09:10Z","timestamp":1789106950598,"version":"build-2803163510"},"publisher-location":"Cham","reference-count":75,"publisher":"Springer Nature Switzerland","isbn-type":[{"value":"9783032377173","type":"print"},{"value":"9783032377180","type":"electronic"}],"license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.springernature.com\/gp\/researchers\/text-and-data-mining"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.springernature.com\/gp\/researchers\/text-and-data-mining"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026]]},"DOI":"10.1007\/978-3-032-37718-0_3","type":"book-chapter","created":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T05:55:48Z","timestamp":1789106148000},"page":"19-37","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Guardrail-Agnostic Societal Bias Evaluation in\u00a0Large Vision-Language Models"],"prefix":"10.1007","author":[{"given":"Yusuke","family":"Hirota","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael Ross","family":"Boone","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arun George","family":"Zachariah","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jibin Rajan","family":"Varghese","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu-Chiang Frank","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Boyi","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ryo","family":"Hachiuma","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,9,12]]},"reference":[{"key":"3_CR1","unstructured":"Alabdulmohsin, I., Wang, X., Steiner, A.P., Goyal, P., D\u2019Amour, A., Zhai, X.: Clip the bias: how useful is balancing data in multimodal learning? In: ICLR (2024)"},{"key":"3_CR2","unstructured":"Anthropic: Claude 3.7 sonnet system card (2025)"},{"key":"3_CR3","unstructured":"Bai, S., et\u00a0al.: Qwen2.5-vl technical report. arXiv preprint arXiv:2502.13923 (2025)"},{"key":"3_CR4","unstructured":"Bai, Y., et\u00a0al.: Constitutional ai: harmlessness from ai feedback. arXiv preprint arXiv:2212.08073 (2022)"},{"key":"3_CR5","doi-asserted-by":"crossref","unstructured":"Balasubramanian, S., Basu, S., Feizi, S.: A closer look at bias and chain-of-thought faithfulness of large (vision) language models. arXiv preprint arXiv:2505.23945 (2025)","DOI":"10.18653\/v1\/2025.findings-emnlp.723"},{"key":"3_CR6","doi-asserted-by":"crossref","unstructured":"Berg, H., et al.: A prompt array keeps the bias away: debiasing vision-language models with adversarial learning. In: AACL (2022)","DOI":"10.18653\/v1\/2022.aacl-main.61"},{"key":"3_CR7","doi-asserted-by":"crossref","unstructured":"Cheng, M., Durmus, E., Jurafsky, D.: Marked personas: using natural language prompts to measure stereotypes in language models. In: ACL (2023)","DOI":"10.18653\/v1\/2023.acl-long.84"},{"key":"3_CR8","unstructured":"Chi, J., et al.: Llama guard 3 vision: safeguarding human-ai image understanding conversations. arXiv preprint arXiv:2411.10414 (2024)"},{"key":"3_CR9","unstructured":"Dehdashtian, S., Wang, L., Boddeti, V.N.: Fairerclip: debiasing clip\u2019s zero-shot predictions using functions in rkhss. In: ICLR (2024)"},{"key":"3_CR10","doi-asserted-by":"crossref","unstructured":"Deitke, M., et al.: Molmo and pixmo: open weights and open data for state-of-the-art vision-language models. In: CVPR (2025)","DOI":"10.1109\/CVPR52734.2025.00018"},{"key":"3_CR11","unstructured":"Ding, Y., Li, L., Cao, B., Shao, J.: Rethinking bottlenecks in safety fine-tuning of vision language models. In: ICLR (2026)"},{"key":"3_CR12","doi-asserted-by":"crossref","unstructured":"Fraser, K.C., Kiritchenko, S.: Examining gender and racial bias in large vision-language models using a novel dataset of parallel images. In: EACL (2024)","DOI":"10.18653\/v1\/2024.eacl-long.41"},{"key":"3_CR13","doi-asserted-by":"crossref","unstructured":"Garcia, N., Hirota, Y., Wu, Y., Nakashima, Y.: Uncurated image-text datasets: shedding light on demographic bias. In: CVPR (2023)","DOI":"10.1109\/CVPR52729.2023.00672"},{"key":"3_CR14","unstructured":"Girrbach, L., Huang, Y., Alaniz, S., Darrell, T., Akata, Z.: Revealing and reducing gender biases in vision and language assistants (vlas). In: ICLR (2025)"},{"key":"3_CR15","doi-asserted-by":"crossref","unstructured":"Gulati, A., D\u2019Inc\u00e0, M., Sebe, N., Lepri, B., Oliver, N.: Beauty and the bias: exploring the impact of attractiveness on multimodal large language models. arXiv preprint arXiv:2504.16104 (2025)","DOI":"10.1609\/aies.v8i2.36619"},{"key":"3_CR16","doi-asserted-by":"crossref","unstructured":"Gustafson, L., et al.: Facet: fairness in computer vision evaluation benchmark. In: ICCV (2023)","DOI":"10.1109\/ICCV51070.2023.01863"},{"key":"3_CR17","doi-asserted-by":"crossref","unstructured":"Harvey, E., et al.: Understanding and meeting practitioner needs when measuring representational harms caused by llm-based systems. In: Findings of ACL (2025)","DOI":"10.18653\/v1\/2025.findings-acl.947"},{"key":"3_CR18","doi-asserted-by":"crossref","unstructured":"Hausladen, C.I., Knott, M., Camerer, C.F., Perona, P.: Social perception of faces in a vision-language model. In: FAccT (2025)","DOI":"10.1145\/3715275.3732041"},{"key":"3_CR19","unstructured":"Hendrycks, D., et al.: Measuring massive multitask language understanding. In: ICLR (2021)"},{"key":"3_CR20","unstructured":"Hirota, Y., Chen, M.H., Wang, C.Y., Nakashima, Y., Wang, Y.C.F., Hachiuma, R.: Saner: annotation-free societal attribute neutralizer for debiasing clip. In: ICLR (2025)"},{"key":"3_CR21","doi-asserted-by":"crossref","unstructured":"Hirota, Y., et al.: Bias in gender bias benchmarks: how spurious features distort evaluation. In: ICCV (2025)","DOI":"10.1109\/ICCV51701.2025.00808"},{"key":"3_CR22","doi-asserted-by":"crossref","unstructured":"Hirota, Y., Nakashima, Y., Garcia, N.: Quantifying societal bias amplification in image captioning. In: CVPR (2022)","DOI":"10.1109\/CVPR52688.2022.01309"},{"key":"3_CR23","doi-asserted-by":"crossref","unstructured":"Howard, P., Bhiwandiwalla, A., Fraser, K.C., Kiritchenko, S.: Uncovering bias in large vision-language models with counterfactuals. In: NAACL (2025)","DOI":"10.18653\/v1\/2025.naacl-long.305"},{"key":"3_CR24","doi-asserted-by":"crossref","unstructured":"Howard, P., Madasu, A., Le, T., Moreno, G.L., Bhiwandiwalla, A., Lal, V.: Probing and mitigating intersectional social biases in vision-language models with counterfactual examples. In: CVPR (2024)","DOI":"10.1109\/CVPR52733.2024.01138"},{"key":"3_CR25","doi-asserted-by":"crossref","unstructured":"Huang, J.T., Qin, J., Zhang, J., Yuan, Y., Wang, W., Zhao, J.: Visbias: measuring explicit and implicit social biases in vision language models. In: EMNLP (2025)","DOI":"10.18653\/v1\/2025.emnlp-main.908"},{"key":"3_CR26","unstructured":"Hurst, A., et al.: Gpt-4o system card. arXiv preprint arXiv:2410.21276 (2024)"},{"key":"3_CR27","doi-asserted-by":"crossref","unstructured":"Jang, T., Jung, H., Wang, X.: Target bias is all you need: zero-shot debiasing of vision-language models with bias corpus. In: ICCV (2025)","DOI":"10.1109\/ICCV51701.2025.00188"},{"key":"3_CR28","unstructured":"Ji, H., Ke, P., Hu, Z., Zhang, R., Huang, M.: Tailoring language generation models under total variation distance. In: ICLR (2023)"},{"key":"3_CR29","unstructured":"Ji, Z., Jia, Y., Gao, S., Yue, Y.: Interpreting social bias in lvlms via information flow analysis and multi-round dialogue evaluation. arXiv preprint arXiv:2505.21106 (2025)"},{"key":"3_CR30","doi-asserted-by":"crossref","unstructured":"Jiang, Y., Li, Z., Shen, X., Liu, Y., Backes, M., Zhang, Y.: Modscan: measuring stereotypical bias in large vision-language models from vision and language modalities. In: EMNLP (2024)","DOI":"10.18653\/v1\/2024.emnlp-main.713"},{"key":"3_CR31","doi-asserted-by":"crossref","unstructured":"Kantharuban, A., Milbauer, J., Sap, M., Strubell, E., Neubig, G.: Stereotype or personalization? User identity biases chatbot recommendations. In: Findings of ACL (2025)","DOI":"10.18653\/v1\/2025.findings-acl.1254"},{"key":"3_CR32","doi-asserted-by":"crossref","unstructured":"Karkkainen, K., Joo, J.: Fairface: face attribute dataset for balanced race, gender, and age for bias measurement and mitigation. In: WACV (2021)","DOI":"10.1109\/WACV48630.2021.00159"},{"key":"3_CR33","unstructured":"Kim, E., et\u00a0al.: World in a frame: understanding culture mixing as a new challenge for vision-language models. In: CVPR (2026)"},{"key":"3_CR34","unstructured":"Kusner, M.J., Loftus, J., Russell, C., Silva, R.: Counterfactual fairness. In: NeurIPS (2017)"},{"key":"3_CR35","unstructured":"Li, B., et\u00a0al.: Llava-onevision: easy visual task transfer. TMLR (2024)"},{"key":"3_CR36","doi-asserted-by":"crossref","unstructured":"Li, Y., Shirado, H., Das, S.: Actions speak louder than words: agent decisions reveal implicit biases in language models. In: FAccT (2025)","DOI":"10.1145\/3715275.3732212"},{"key":"3_CR37","doi-asserted-by":"crossref","unstructured":"Liu, H., Li, C., Li, Y., Lee, Y.J.: Improved baselines with visual instruction tuning. In: CVPR (2024)","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"3_CR38","doi-asserted-by":"crossref","unstructured":"Liu, Q., et al.: Unraveling and mitigating safety alignment degradation of vision-language models. In: Findings of ACL (2025)","DOI":"10.18653\/v1\/2025.findings-acl.186"},{"key":"3_CR39","doi-asserted-by":"crossref","unstructured":"Liu, Y., et\u00a0al.: Guardreasoner-vl: safeguarding vlms via reinforced reasoning. In: NeurIPS (2025)","DOI":"10.52202\/085713-0977"},{"key":"3_CR40","doi-asserted-by":"crossref","unstructured":"Malik, S., Abdullah, H.M., Saha, S., Sheth, A.: Ask me again differently: gras for measuring bias in vision language models on gender, race, age, and skin tone. arXiv preprint arXiv:2508.18989 (2025)","DOI":"10.18653\/v1\/2026.findings-eacl.123"},{"key":"3_CR41","doi-asserted-by":"crossref","unstructured":"Markov, T., et al.: A holistic approach to undesired content detection in the real world. In: AAAI (2023)","DOI":"10.1609\/aaai.v37i12.26752"},{"key":"3_CR42","doi-asserted-by":"crossref","unstructured":"Meister, N., Zhao, D., Wang, A., Ramaswamy, V.V., Fong, R., Russakovsky, O.: Gender artifacts in visual datasets. In: ICCV (2023)","DOI":"10.1109\/ICCV51070.2023.00446"},{"key":"3_CR43","unstructured":"Narnaware, V., Vayani, A., Gupta, R., Swetha, S., Shah, M.: Sb-bench: stereotype bias benchmark for large multimodal models. arXiv preprint arXiv:2502.08779 (2025)"},{"key":"3_CR44","unstructured":"OpenAI: Gpt-5 system card (2025). https:\/\/openai.com\/index\/gpt-5-system-card\/. Accessed Sept 2025"},{"key":"3_CR45","unstructured":"Qi, X., et al.: Safety alignment should be made more than just a few tokens deep. In: ICLR (2025)"},{"key":"3_CR46","doi-asserted-by":"crossref","unstructured":"Raj, C., Mukherjee, A., Caliskan, A., Anastasopoulos, A., Zhu, Z.: Biasdora: exploring hidden biased associations in vision-language models. In: EMNLP Findings (2024)","DOI":"10.18653\/v1\/2024.findings-emnlp.611"},{"key":"3_CR47","doi-asserted-by":"crossref","unstructured":"Raj, C., Wei, B., Caliskan, A., Anastasopoulos, A., Zhu, Z.: Vignette: socially grounded bias evaluation for vision-language models. arXiv preprint arXiv:2505.22897(2025)","DOI":"10.18653\/v1\/2026.acl-long.712"},{"key":"3_CR48","doi-asserted-by":"crossref","unstructured":"Ratzlaff, N., Olson, M.L., Hinck, M., Tseng, S.Y., Lal, V., Howard, P.: Debias your large multi-modal model at test-time with non-contrastive visual attribute steering. In: ICCV (2025)","DOI":"10.1109\/ICCVW69036.2025.00651"},{"key":"3_CR49","doi-asserted-by":"crossref","unstructured":"Raza, S., et al.: Humanibench: a human-centric framework for large multimodal models evaluation. arXiv preprint arXiv:2505.11454 (2025)","DOI":"10.1145\/3845999"},{"key":"3_CR50","doi-asserted-by":"crossref","unstructured":"Salewski, L., Alaniz, S., Rio-Torto, I., Schulz, E., Akata, Z.: In-context impersonation reveals large language models\u2019 strengths and biases. In: NeurIPS (2023)","DOI":"10.52202\/075280-3152"},{"key":"3_CR51","doi-asserted-by":"crossref","unstructured":"Sathe, A., Jain, P., Sitaram, S.: A unified framework and dataset for assessing societal bias in vision-language models. In: EMNLP Findings (2024)","DOI":"10.18653\/v1\/2024.findings-emnlp.66"},{"key":"3_CR52","doi-asserted-by":"crossref","unstructured":"Seth, A., Hemani, M., Agarwal, C.: Dear: debiasing vision-language models with additive residuals. In: CVPR (2023)","DOI":"10.1109\/CVPR52729.2023.00659"},{"key":"3_CR53","unstructured":"Team, G., et al.: Gemma 3 technical report. arXiv preprint arXiv:2503.19786 (2025)"},{"key":"3_CR54","doi-asserted-by":"crossref","unstructured":"Van Handel, R.: Probability in high dimension. Technical Report (2014)","DOI":"10.21236\/ADA623999"},{"key":"3_CR55","doi-asserted-by":"crossref","unstructured":"Wang, A., Narayanan, A., Russakovsky, O.: REVISE: a tool for measuring and mitigating bias in visual datasets. In: ECCV (2020)","DOI":"10.1007\/978-3-030-58580-8_43"},{"key":"3_CR56","doi-asserted-by":"crossref","unstructured":"Wang, A., Phan, M., Ho, D.E., Koyejo, S.: Fairness through difference awareness: measuring desired group discrimination in llms. In: ACL (2025)","DOI":"10.18653\/v1\/2025.acl-long.341"},{"key":"3_CR57","unstructured":"Wang, P., et\u00a0al.: Qwen2-vl: enhancing vision-language model\u2019s perception of the world at any resolution. arXiv preprint arXiv:2409.12191 (2024)"},{"key":"3_CR58","unstructured":"Wang, S., et al.: Vlbiasbench: a comprehensive benchmark for evaluating bias in large vision-language model. arXiv preprint arXiv:2406.14194 (2024)"},{"key":"3_CR59","doi-asserted-by":"crossref","unstructured":"Wang, T., Zhao, J., Yatskar, M., Chang, K.W., Ordonez, V.: Balanced datasets are not enough: estimating and mitigating gender bias in deep image representations. In: ICCV (2019)","DOI":"10.1109\/ICCV.2019.00541"},{"key":"3_CR60","unstructured":"Wang, W., et al.: Internvl3. 5: advancing open-source multimodal models in versatility, reasoning, and efficiency. arXiv preprint arXiv:2508.18265 (2025)"},{"key":"3_CR61","doi-asserted-by":"crossref","unstructured":"Wang, X., Yi, X., Jiang, H., Zhou, S., Wei, Z., Xie, X.: Tovilag: your visual-language generative model is also an evildoer. In: EMNLP (2023)","DOI":"10.18653\/v1\/2023.emnlp-main.213"},{"key":"3_CR62","doi-asserted-by":"crossref","unstructured":"Wang, Y., Liu, X., Li, Y., Chen, M., Xiao, C.: Adashield: safeguarding multimodal large language models from structure-based attack via adaptive shield prompting. In: ECCV (2024)","DOI":"10.1007\/978-3-031-72661-3_5"},{"key":"3_CR63","doi-asserted-by":"crossref","unstructured":"Wu, X., Wang, Y., Wu, H.T., Tao, Z., Fang, Y.: Evaluating fairness in large vision-language models across diverse demographic attributes and prompts. In: EMNLP (2025)","DOI":"10.18653\/v1\/2025.findings-emnlp.1251"},{"key":"3_CR64","doi-asserted-by":"crossref","unstructured":"Xiang, A., et\u00a0al.: Fair human-centric image dataset for ethical ai benchmarking. Nature (2025)","DOI":"10.1038\/s41586-025-09716-2"},{"key":"3_CR65","doi-asserted-by":"crossref","unstructured":"Xiao, Y., et al.: Genderbias-vl: benchmarking gender bias in vision language models via counterfactual probing. IJCV (2025)","DOI":"10.1007\/s11263-025-02556-7"},{"key":"3_CR66","unstructured":"Xu, Y., Wang, W.: From individuals to interactions: benchmarking gender bias in multimodal large language models from the lens of social relationship. arXiv preprint arXiv:2506.23101 (2025)"},{"key":"3_CR67","unstructured":"Yang, A., et\u00a0al.: Qwen3 technical report. arXiv preprint arXiv:2505.09388 (2025)"},{"key":"3_CR68","doi-asserted-by":"crossref","unstructured":"Yue, X., et\u00a0al.: Mmmu: a massive multi-discipline multimodal understanding and reasoning benchmark for expert agi. In: CVPR (2024)","DOI":"10.1109\/CVPR52733.2024.00913"},{"key":"3_CR69","doi-asserted-by":"crossref","unstructured":"Zhang, H., Guo, Y., Kankanhalli, M.: Joint vision-language social bias removal for clip. In: CVPR (2025)","DOI":"10.1109\/CVPR52734.2025.00401"},{"key":"3_CR70","doi-asserted-by":"crossref","unstructured":"Zhang, Y., et\u00a0al.: Spa-vl: a comprehensive safety preference alignment dataset for vision language models. In: CVPR (2025)","DOI":"10.1109\/CVPR52734.2025.01850"},{"key":"3_CR71","doi-asserted-by":"crossref","unstructured":"Zhao, J., Wang, T., Yatskar, M., Ordonez, V., Chang, K.W.: Men also like shopping: reducing gender bias amplification using corpus-level constraints. In: EMNLP (2017)","DOI":"10.18653\/v1\/D17-1323"},{"key":"3_CR72","doi-asserted-by":"crossref","unstructured":"Zhao, S., et al.: Jailbreaking multimodal large language models via shuffle inconsistency. In: ICCV (2025)","DOI":"10.1109\/ICCV51701.2025.00198"},{"key":"3_CR73","unstructured":"Zhao, Z., Yamasaki, T.: Bias beyond demographics: probing decision boundaries in black-box lvlms via counterfactual vqa. arXiv preprint arXiv:2508.03079 (2025)"},{"key":"3_CR74","unstructured":"Zhu, J., et al.: Internvl3: exploring advanced training and test-time recipes for open-source multimodal models. arXiv preprint arXiv:2504.10479 (2025)"},{"key":"3_CR75","unstructured":"Zong, Y., Bohdal, O., Yu, T., Yang, Y., Hospedales, T.: Safety fine-tuning at (almost) no cost: a baseline for vision large language models. In: ICML (2024)"}],"container-title":["Lecture Notes in Computer Science","Computer Vision \u2013 ECCV 2026"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/978-3-032-37718-0_3","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T05:56:12Z","timestamp":1789106172000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/978-3-032-37718-0_3"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"ISBN":["9783032377173","9783032377180"],"references-count":75,"URL":"https:\/\/doi.org\/10.1007\/978-3-032-37718-0_3","relation":{},"ISSN":["0302-9743","1611-3349"],"issn-type":[{"value":"0302-9743","type":"print"},{"value":"1611-3349","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]},"assertion":[{"value":"12 September 2026","order":1,"name":"first_online","label":"First Online","group":{"name":"ChapterHistory","label":"Chapter History"}},{"value":"ECCV","order":1,"name":"conference_acronym","label":"Conference Acronym","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"European Conference on Computer Vision","order":2,"name":"conference_name","label":"Conference Name","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"Malm\u00f6","order":3,"name":"conference_city","label":"Conference City","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"Sweden","order":4,"name":"conference_country","label":"Conference Country","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"2026","order":5,"name":"conference_year","label":"Conference Year","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"8 September 2026","order":7,"name":"conference_start_date","label":"Conference Start Date","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"12 September 2026","order":8,"name":"conference_end_date","label":"Conference End Date","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"19","order":9,"name":"conference_number","label":"Conference Number","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"eccv2026","order":10,"name":"conference_id","label":"Conference ID","group":{"name":"ConferenceInfo","label":"Conference Information"}},{"value":"https:\/\/eccv.ecva.net\/","order":11,"name":"conference_url","label":"Conference URL","group":{"name":"ConferenceInfo","label":"Conference Information"}}]}}