Update paper links and citation to published EMNLP Findings version

CITATION.cff

@@ -53,11 +53,17 @@ preferred-citation:
     LLMs Can Predict Failure Risk, But Struggle to Predict Which Collaboration
     Protocol Pays Off: Cost-Aware Protocol Routing Across Reasoning Tasks
   year: 2026
-  collection-title: "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP)"
+  month: 10
+  start: 26995
+  end: 27016
+  doi: "10.18653/2026.findings-emnlp.1275"
+  publisher:
+    name: "Association for Computational Linguistics"
+  collection-title: "Findings of the Association for Computational Linguistics: EMNLP 2026"
   identifiers:
     - type: other
       value: "arXiv:2608.14927"
-  url: "https://arxiv.org/abs/2608.14927"
+  url: "https://aclanthology.org/2026.findings-emnlp.1275/"
   authors:
     - family-names: Yang
       given-names: Chih-Hsuan

README.md

@@ -106,9 +106,9 @@ predict which setup is worth paying for.
 It can predict *whether it will fail*. It cannot predict *which collaboration
 protocol will fix the failure*. That gap is what this dataset is for.
 
-This is the data release for the EMNLP 2026 paper
+This is the data release for the Findings of EMNLP 2026 paper
 [*LLMs Can Predict Failure Risk, But Struggle to Predict Which Collaboration
-Protocol Pays Off*](https://arxiv.org/abs/2608.14927).
+Protocol Pays Off*](https://aclanthology.org/2026.findings-emnlp.1275/).
 
 ## The four protocols
 
@@ -464,7 +464,8 @@ probe.
 
 ## Links
 
-- Paper: <https://arxiv.org/abs/2608.14927>
+- Paper (ACL Anthology): <https://aclanthology.org/2026.findings-emnlp.1275/>
+- Official PDF: <https://aclanthology.org/2026.findings-emnlp.1275.pdf>
 - Code: <https://github.com/ChihHsuan-Yang/EMNLP_Cost-Aware-Protocol-Routing>
 - Project site: <https://chihhsuan-yang.github.io/EMNLP_Cost-Aware-Protocol-Routing/>
 - Model card: <https://huggingface.co/AgentsSci/EMNLP_Cost-Aware-Protocol-Routing>
@@ -472,18 +473,28 @@ probe.
 ## Citation
 
 ```bibtex
-@inproceedings{yang2026costaware,
-  title     = {LLMs Can Predict Failure Risk, But Struggle to Predict Which
-               Collaboration Protocol Pays Off: Cost-Aware Protocol Routing
-               Across Reasoning Tasks},
-  author    = {Yang, Chih-Hsuan and Jiang, Jingyan and Yang, Cheng-Hau and
-               Vasudevan, Vikram and Zheng, Huihuo and Vishwanath, Venkatram and
-               Thakur, Rajeev},
-  booktitle = {Proceedings of the 2026 Conference on Empirical Methods in
-               Natural Language Processing (EMNLP)},
-  year      = {2026},
-  eprint    = {2608.14927},
-  archivePrefix = {arXiv}
+@inproceedings{yang-etal-2026-llms-predict,
+    title = "{LLM}s Can Predict Failure Risk, But Struggle to Predict Which Collaboration Protocol Pays Off: Cost-Aware Protocol Routing Across Reasoning Tasks",
+    author = "Yang, Chih-Hsuan  and
+      Jiang, Jingyan  and
+      Yang, Cheng-Hau  and
+      Vasudevan, Vikram  and
+      Zheng, Huihuo  and
+      Vishwanath, Venkatram  and
+      Thakur, Rajeev",
+    editor = "Lee, Hung-Yi  and
+      Dev, Sunipa  and
+      Li, Jessy  and
+      Elliott, Desmond",
+    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
+    month = oct,
+    year = "2026",
+    address = "Budapest, Hungary",
+    publisher = "Association for Computational Linguistics",
+    url = "https://aclanthology.org/2026.findings-emnlp.1275/",
+    doi = "10.18653/2026.findings-emnlp.1275",
+    pages = "26995--27016",
+    ISBN = "979-8-89176-462-0"
 }
 ```
 

registry/checksums.sha256

@@ -1,7 +1,7 @@
-be06f3a4e4d079f7a34cf319636f88050fca605ca7c3f02c595e36daabd214c4  CITATION.cff
+f29c71a117aa8b28fb25e453b4953c564a7e47807c307fc771acde11414044a7  CITATION.cff
 fe34eb7ac34537c21249e1ff6c9cf7b863eb611f844c0cb212ea358576e32887  LICENSE
 37d5e9d642d6684659d096bd8ebb5ea2de7ccc487fcdb2e4479dfa073c322f23  NOTICE
-6d5a8469900a3e9659123d0047874491b2b5019b52cc2eedb47b80a3e8615a76  README.md
+95006aa520ae00a389835868e0a3205aa3b30794fbcaa6888ac0427e65b576f5  README.md
 d27d1195b4be7c75016ca83096d658548d1127de709c11061c733dc940e7c6e9  TODO.md
 aad9af7b8ae8cbee7a7d73ecb7393cbb5adcb6ded6d5ec924dc8d1487e34c045  data/aggregate/failure_and_protocol_value_targets.csv
 ff064f045fcfab7e352d58f56070d98c3d1461e6cf44d3dfa0c5b108713c3fa9  data/aggregate/heldout_router_evaluation.csv
@@ -39,6 +39,6 @@ abcfeff67504846bd853c573170fa2650761ebc6356b241b1149bffa34417c01  docs/reconstru
 1a97156270d9fcec83019bef45e7ce9f7eaf61befd1cbcc736f6f27800bb2187  registry/experiments.csv
 0b27a71a1bc8ce61efeadec7722cabc566286e5a335cf213fa21b05fcdc4d3c4  registry/models.csv
 e263c1473f08b37726726ee8b7c206f6799fa79e4fd1a128ee121ec3197c8ca5  registry/protocols.csv
-09b56c1583699d9ca6915a5840cc57e0e0a882ed7e320c104f632140098767bc  registry/release_manifest.json
+d42005ab9f511b20603a0feda17398a4ba4f20a10f1f1f3c324582f2a4817869  registry/release_manifest.json
 3aea6a2d356a1809b69c3e8b0aaf9c54dfe05cbf73d5e2368f06e9f47c90aa70  registry/schema.json
 8e69d6584f3d059969eeb8a6515aa621305f1e223ebd4ffc501731b9f14c20b2  validate.py

registry/release_manifest.json

@@ -4,12 +4,17 @@
   "generated_utc": "2026-09-23T23:53:48.235395+00:00",
   "paper": {
     "title": "LLMs Can Predict Failure Risk, But Struggle to Predict Which Collaboration Protocol Pays Off: Cost-Aware Protocol Routing Across Reasoning Tasks",
-    "venue": "EMNLP 2026",
+    "venue": "Findings of the Association for Computational Linguistics: EMNLP 2026",
     "arxiv": "https://arxiv.org/abs/2608.14927",
     "code": "https://github.com/ChihHsuan-Yang/EMNLP_Cost-Aware-Protocol-Routing",
     "site": "https://chihhsuan-yang.github.io/EMNLP_Cost-Aware-Protocol-Routing/",
     "model_card": "https://huggingface.co/AgentsSci/EMNLP_Cost-Aware-Protocol-Routing",
-    "contact": "bellayang@anl.gov"
+    "contact": "bellayang@anl.gov",
+    "url": "https://aclanthology.org/2026.findings-emnlp.1275/",
+    "pdf": "https://aclanthology.org/2026.findings-emnlp.1275.pdf",
+    "doi": "10.18653/2026.findings-emnlp.1275",
+    "pages": "26995–27016",
+    "publication_date": "2026-10"
   },
   "prompt_versions": {
     "six_setting_confidence_probe": "confidence-probe-v3",
@@ -264,5 +269,6 @@
     },
     "caveat": "Every license above was read from a file in the project tree, not fetched from upstream at audit time. Re-verify before public upload."
   },
-  "acknowledgment": "This research used resources of the Argonne Leadership Computing Facility, a U.S. Department of Energy (DOE) Office of Science user facility at Argonne National Laboratory (ANL) operated under Contract No. DE-AC02-06CH11357."
-}
+  "acknowledgment": "This research used resources of the Argonne Leadership Computing Facility, a U.S. Department of Energy (DOE) Office of Science user facility at Argonne National Laboratory (ANL) operated under Contract No. DE-AC02-06CH11357.",
+  "publication_metadata_updated": "2026-10-30"
+}