6 papers
Tokengeist: Multi-Turn Attribution Tracing in Agentic Conversations
Jessica Tang, Shraddha Barke, Sharad Agarwal
When a language model produces a response in a multi-turn conversation, which tokens from prior turns shaped that answer, and how did those dependencies propagate across prior turn…
Willful Disobedience: Automatically Detecting Failures in Agentic Traces
Reshabh K Sharma, Shraddha Barke, Benjamin Zorn
AI agents are increasingly embedded in real software systems, where they execute multi-step workflows through multi-turn dialogue, tool invocations, and intermediate decisions. The…
PromptPex: Automatic Test Generation for Language Model Prompts
Reshabh K Sharma, Jonathan De Halleux, Shraddha Barke +2
Large language models (LLMs) are being used in many applications and prompts for these models are integrated into software applications as code-like artifacts. These prompts behave…
AgentRx: Diagnosing AI Agent Failures from Execution Trajectories
Shraddha Barke, Arnav Goyal, Alind Khare +3
AI agents often fail in ways that are difficult to localize because executions are probabilistic, long-horizon, multi-agent, and mediated by noisy tool outputs. We address this gap…
What's in a Proof? Analyzing Expert Proof-Writing Processes in F* and Verus
Rijul Jain, Shraddha Barke, Gabriel Ebner +3
Proof-oriented programming languages (POPLs) empower developers to write code alongside formal correctness proofs, providing formal guarantees that the code adheres to specified re…
HYSYNTH: Context-Free LLM Approximation for Guiding Program Synthesis
Shraddha Barke, Emmanuel Anaya Gonzalez, Saketh Ram Kasibatla +2
Many structured prediction and reasoning tasks can be framed as program synthesis problems, where the goal is to generate a program in a domain-specific language (DSL) that transfo…