{"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/websailor-navigating-super-human-reasoning","title":"WebSailor: Navigating Super-human Reasoning for Web Agent","arxiv_id":"2507.02592","date":"2025-07-03","proceeding":null,"authors":["Kuan Li","Zhongwang Zhang","Huifeng Yin","Liwen Zhang","Litu Ou","Jialong Wu","Wenbiao Yin","Baixuan Li","Zhengwei Tao","Xinyu Wang","Weizhou Shen","Junkai Zhang","Dingchu Zhang","Xixi Wu","Yong Jiang","Ming Yan","Pengjun Xie","Fei Huang","Jingren Zhou"],"abstract":"Transcending human cognitive limitations represents a critical frontier in LLM training. Proprietary agentic systems like DeepResearch have demonstrated superhuman capabilities on extremely complex information-seeking benchmarks such as BrowseComp, a feat previously unattainable. We posit that their success hinges on a sophisticated reasoning pattern absent in open-source models: the ability to systematically reduce extreme uncertainty when navigating vast information landscapes. Based on this insight, we introduce WebSailor, a complete post-training methodology designed to instill this crucial capability. Our approach involves generating novel, high-uncertainty tasks through structured sampling and information obfuscation, RFT cold start, and an efficient agentic RL training algorithm, Duplicating Sampling Policy Optimization (DUPO). With this integrated pipeline, WebSailor significantly outperforms all opensource agents in complex information-seeking tasks, matching proprietary agents' performance and closing the capability gap.","url_abs":"https://arxiv.org/abs/2507.02592v1","url_pdf":"https://arxiv.org/pdf/2507.02592v1.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":"websailor-navigating-super-human-reasoning","repo_url":"https://github.com/alibaba-nlp/webagent","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"websailor-navigating-super-human-reasoning","repo_url":"https://github.com/SnailDev/github-hot-hub","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"websailor-navigating-super-human-reasoning","repo_url":"https://github.com/alibaba-nlp/webwalker","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"websailor-navigating-super-human-reasoning","repo_url":"https://github.com/kun-g/Scraping-Github-trending","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2507.02592","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2507.02592"}},"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/alibaba-nlp/webwalker","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/SnailDev/github-hot-hub","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/alibaba-nlp/webagent","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kun-g/Scraping-Github-trending","reach":{"status":"ok"}}],"summary":{"unverified":22},"by_repo_kind":{"official":{"samples":12,"ran":0,"repositories":1},"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":"cc601f60e63e7a5c","entry":"call_llm","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/NestBrowse/utils.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/NestBrowse/utils.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":"cc601f60e63e7a5c"}},{"code_sha256_prefix":"ffcb6da25872786e","entry":"call_llm","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/ParallelMuse/functionality_specified_partial_rollout.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/ParallelMuse/functionality_specified_partial_rollout.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":"ffcb6da25872786e"}},{"code_sha256_prefix":"09d32bbfdc4d6cf2","entry":"call_llm_judge","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"evaluation/evaluate_deepsearch_official.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/evaluation/evaluate_deepsearch_official.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":"09d32bbfdc4d6cf2"}},{"code_sha256_prefix":"8605222f211cb27e","entry":"call_resum_server","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/WebResummer/src/summary_utils.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/WebResummer/src/summary_utils.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":"8605222f211cb27e"}},{"code_sha256_prefix":"df838ce4c63f0ea4","entry":"call_tool","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/NestBrowse/infer_async_nestbrowse.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/NestBrowse/infer_async_nestbrowse.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":"df838ce4c63f0ea4"}},{"code_sha256_prefix":"c517da931a712d31","entry":"call_tool","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/ParallelMuse/functionality_specified_partial_rollout.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/ParallelMuse/functionality_specified_partial_rollout.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":"c517da931a712d31"}},{"code_sha256_prefix":"c2e42aa986785080","entry":"clean_markdown","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"WebAgent/WebWalker/src/utils.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/WebAgent/WebWalker/src/utils.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":"c2e42aa986785080"}},{"code_sha256_prefix":"ef0eff9da9ccbda2","entry":"cluster_by_question","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/ParallelMuse/compressed_reasoning_aggregation.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/ParallelMuse/compressed_reasoning_aggregation.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":"ef0eff9da9ccbda2"}},{"code_sha256_prefix":"23aacba017665d93","entry":"count_tokens","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/NestBrowse/utils.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/NestBrowse/utils.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":"23aacba017665d93"}},{"code_sha256_prefix":"079a31f347eec894","entry":"extract_answer","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"evaluation/evaluate_hle_official.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/evaluation/evaluate_hle_official.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":"079a31f347eec894"}},{"code_sha256_prefix":"a4ee90a1bd5be041","entry":"extract_response","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"evaluation/evaluate_hle_official.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/evaluation/evaluate_hle_official.py","link_basis":"plan_row","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":"a4ee90a1bd5be041"}},{"code_sha256_prefix":"6f7ac5ce08066791","entry":"extract_tags","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/AgentFold/infer.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/AgentFold/infer.py","link_basis":"plan_row","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":"6f7ac5ce08066791"}},{"code_sha256_prefix":"e6b851ca45ef45b8","entry":"gemini_api","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"WebAgent/WebWalker/src/rag_system.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/WebAgent/WebWalker/src/rag_system.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":"e6b851ca45ef45b8"}},{"code_sha256_prefix":"f78af907d3807fbc","entry":"get_llm_response","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/ParallelMuse/compressed_reasoning_aggregation.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/ParallelMuse/compressed_reasoning_aggregation.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":"f78af907d3807fbc"}},{"code_sha256_prefix":"c06aada9aaf8d772","entry":"get_llm_response_nonstream","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/AgentFold/infer.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/AgentFold/infer.py","link_basis":"plan_row","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":"c06aada9aaf8d772"}},{"code_sha256_prefix":"11c85a122a8418ae","entry":"is_correct_judgement","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"evaluation/evaluate_deepsearch_official.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/evaluation/evaluate_deepsearch_official.py","link_basis":"plan_row","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":"11c85a122a8418ae"}},{"code_sha256_prefix":"e587e0fad56e82f2","entry":"load_jsonl","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"evaluation/evaluate_hle_official.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/evaluation/evaluate_hle_official.py","link_basis":"plan_row","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":"e587e0fad56e82f2"}},{"code_sha256_prefix":"149cf8d65d996603","entry":"o1_api","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"WebAgent/WebWalker/src/rag_system.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/WebAgent/WebWalker/src/rag_system.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":"149cf8d65d996603"}},{"code_sha256_prefix":"d9c80590a339f019","entry":"process_single_round","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"evaluation/evaluate_deepsearch_official.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/evaluation/evaluate_deepsearch_official.py","link_basis":"plan_row","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":"d9c80590a339f019"}},{"code_sha256_prefix":"d48b7faf85d5ebcb","entry":"process_url","repo":"alibaba-nlp/webwalker","repo_kind":"listed","path":"WebAgent/WebWalker/src/utils.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/WebAgent/WebWalker/src/utils.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":"d48b7faf85d5ebcb"}},{"code_sha256_prefix":"ecb8ee265d233ddd","entry":"read_jsonl","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/NestBrowse/utils.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/NestBrowse/utils.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":"ecb8ee265d233ddd"}},{"code_sha256_prefix":"d903fce78a4ff441","entry":"single_round_statistics","repo":"alibaba-nlp/webagent","repo_kind":"official","path":"WebAgent/WebResummer/src/evaluate.py","file_url":"https://github.com/alibaba-nlp/webagent/blob/HEAD/WebAgent/WebResummer/src/evaluate.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":"d903fce78a4ff441"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}