Update 2017-02-01-cv.md

Author: hsanchez

posts/notes/_posts/2017-02-01-cv.md

@@ -171,32 +171,31 @@ hidden: true
         <div class="year">2025</div>
         <div class="item">
             <span class="bold"><a href="https://arxiv.org/abs/2510.03930">Reliable Multi-AI Collaboration</a></span> (Team: SRI) <br />
-            <span class="deemph"> Leading the development of a post-training framework for optimizing model performance for complex software tasks via LLM Chemistry estimation, optimal multi-LLM collaboration, tool integration via MCP, prompt strategies, and probabilistic multi-LLM consensus mechanisms for uncertainty quantification to ensure the reliable, consistent, and correct execution of complex reasoning tasks.
-          </span>
+            <span class="deemph">Developed a post-training framework for multi-LLM collaboration optimization — including Multi-LLM Chemistry estimation and Multi-LLM consensus mechanisms — to improve reliability and consistency across complex reasoning tasks. Designed benchmarks and evaluation methods to measure collective model performance and quantify uncertainty in multi-agent outputs.</span>
         </div>
     </div>
     <div class="entry">
         <!--<div class="year">2024 &ndash; now</div>-->
         <div class="year">2024</div>
         <div class="item">
             <span class="bold"><a href="https://arpa-h.gov/research-and-funding/programs/paradigm">ARPA-H PARADIGM</a></span> (Team: SRI) <br />
-            <span class="deemph">Developing applied multi-LLM workflows for extracting structure from messy healthcare data formats, inferring JSON schemas and constraints, and synthesizing schema-compliant code for cross-device data translation. Built evaluation and failure-analysis methods to profile hallucinated constraints, brittle formatting, and retrieval/grounding failures in domain-specific data workflows.</span>
+            <span class="deemph">Developed multi-LLM (Agentic) workflows to automatically extract structure from messy, domain-specific data formats — inferring JSON schemas and constraints, and synthesizing schema-compliant code for cross-device data translation. Built rigorous evaluation and failure-analysis methods to profile hallucinated constraints, brittle formatting, and retrieval/grounding failures.</span>
         </div>
     </div>
     <div class="entry">
         <!--<div class="year">2024 &ndash; now</div>-->
         <div class="year">2024</div>
         <div class="item">
             <span class="bold"><a href="https://www.iarpa.gov/research-programs/rescind">IARPA ReSCIND</a></span> (Team: SRI, IHMC, TwoSix, UFL, RAD, GMU, ColumbiaU, SimSpace) <br />
-            <span class="deemph">Leading the development of Cyber-Specific Computational Cognitive Models using Self-Supervised Machine Learning, Information Foraging Theory, and Fuzzy Cognitive Maps.</span>
+            <span class="deemph">Developed Cyber-Specific Computational Cognitive Models (C3Ms) using self-supervised machine learning, Information Foraging Theory, and Fuzzy Cognitive Maps to model attacker decision-making in contested cyber environments.</span>
         </div>
     </div>
     <div class="entry">
         <!--<div class="year">2024 &ndash; now</div>-->
         <div class="year">2024</div>
         <div class="item">
             <span class="bold"><a href="https://arpa-h.gov/research-and-funding/programs/digiheals">ARPA-H DIGIHEALS</a></span> (Team: SRI, Confidencial Inc., Homeward Health) <br />
-            <span class="deemph">Advancing the automatic generation of formal representations from natural language using neuro-symbolic AI and retrieval augmented generation (RAG), and leading the development of multi-LLM (Agentic) workflows for healthcare data analysis, formalization, and knowledge composition.</span>
+            <span class="deemph">Developed multi-LLM workflows for healthcare data analysis, formalization, and knowledge composition, and built neurosymbolic AI and RAG-based methods for automatic generation of formal representations from natural language.</span>
         </div>
     </div>
     <div class="heading">Selected Earlier Projects<sup>*</sup></div>
@@ -239,7 +238,7 @@ hidden: true
         <div class="item">
             <span class="bold">SRI International</span> &middot; Menlo Park, CA <br />
             <span class="emph">Senior Computer Scientist</span>, Computer Science Laboratory <br/>
-            <span class="deemph">Researched inference-time optimization for compound software tasks — automated program repair and code generation — developing methods for optimal multi-LLM collaboration, LLM chemistry estimation, and multi-LLM consensus for reliable and consistent collective inference.</span>
+            <span class="deemph">Researched inference-time optimization for compound software tasks — automated program repair and code generation — developing methods for optimal multi-LLM collaboration, Multi-LLM Chemistry estimation, and probabilistic consensus for reliable and consistent collective inference.</span>
         </div>
     </div>
     <div class="entry">