Add BnB code (#410)

Author: TCtower

source/_data/SymbioticLab.bib

@@ -2173,6 +2173,7 @@ @Article{branchandbrowse:arxiv25
   url           = {https://arxiv.org/abs/2510.19838},
   publist_confkey = {arXiv:2510.19838},
   publist_link = {paper || https://arxiv.org/abs/2510.19838},
+  publist_link = {code || https://github.com/SymbioticLab/Branch-and-Browse},
   publist_topic = {Systems + AI},
   publist_abstract = {
 Autonomous web agents powered by large language models (LLMs) show strong potential for performing goal-oriented tasks such as information retrieval, report generation, and online transactions. These agents mark a key step toward practical embodied reasoning in open web environments. However, existing approaches remain limited in reasoning depth and efficiency: vanilla linear methods fail at multi-step reasoning and lack effective backtracking, while other search strategies are coarse-grained and computationally costly. We introduce Branch-and-Browse, a fine-grained web agent framework that unifies structured reasoning-acting, contextual memory, and efficient execution. It (i) employs explicit subtask management with tree-structured exploration for controllable multi-branch reasoning, (ii) bootstraps exploration through efficient web state replay with background reasoning, and (iii) leverages a page action memory to share explored actions within and across sessions. On the WebArena benchmark, Branch-and-Browse achieves a task success rate of 35.8% and reduces execution time by up to 40.4% relative to state-of-the-art methods. These results demonstrate that Branch-and-Browse is a reliable and efficient framework for LLM-based web agents.
@@ -2448,6 +2449,7 @@ @InProceedings{branchandbrowse:acl26
   month         = {July},
   publist_confkey = {ACL'26},
   publist_link = {paper || branchandbrowse-acl26.pdf},
+  publist_link = {code || https://github.com/SymbioticLab/Branch-and-Browse},
   publist_topic = {Systems + AI},
   publist_abstract = {
 Autonomous web agents powered by large language models (LLMs) show strong potential for performing goal-oriented tasks such as information retrieval, report generation, and online transactions. These agents mark a key step toward practical embodied reasoning in open web environments. However, existing approaches remain limited in reasoning depth and efficiency: vanilla linear methods fail at multi-step reasoning and lack effective backtracking, while other search strategies are coarse-grained and computationally costly. We introduce Branch-and-Browse, a fine-grained web agent framework that unifies structured reasoning-acting, contextual memory, and efficient execution. It (i) employs explicit subtask management with tree-structured exploration for controllable multi-branch reasoning, (ii) bootstraps exploration through efficient web state replay with background reasoning, and (iii) leverages a page action memory to share explored actions within and across sessions. On the WebArena benchmark, Branch-and-Browse achieves a task success rate of 35.8% and reduces execution time by up to 40.4% relative to state-of-the-art methods. These results demonstrate that Branch-and-Browse is a reliable and efficient framework for LLM-based web agents.

source/open-source/index.md

@@ -63,6 +63,9 @@ sections:
 
   - title: "Selected Software Artifacts"
     items:
+      - name: "Branch-and-Browse"
+        github_url: "https://github.com/SymbioticLab/Branch-and-Browse"
+        description: "Efficient and controllable web exploration for LLM agents."
       - name: "Kareus"
         github_url: "https://github.com/ml-energy/kareus"
         description: "Train large models with less energy and time."