5 papers
Mind the Cap: Output-Budget Regimes Change the Measured Multilingual Reasoning Gap
Ankit Goyal, Jaideep Ray
Multilingual evaluations report accuracy at a single output-token cap, but languages need different numbers of tokens to express the same content, so the cap is a hidden experiment…
From Agent Failures to Text Policies: What Works and What Breaks
Jaideep Ray, Ankit Goyal
TextGrad improves language-model systems by revising text from feedback. Its core thesis is that natural-language feedback can act as a gradient for optimizing text components with…
Structured Feedback Improves Repair in an LLM Agent Loop
Jaideep Ray, Ankit Goyal
The paper presents VeriHarness, a code‑controlled loop that lets external validators give structured feedback (failure location, observed value, admissible alternatives) to LLM age…
Before the Model Learns the Bug:Fuzzing RLVR Verifiers
Jaideep Ray
Reinforcement learning with verifiable rewards (RLVR) replaces human preference labels with executable reward functions such as math answer checkers, JSON tool-call validators, and…
The Constraint Tax: Measuring Validity-Correctness Tradeoffs in Structured Outputs for Small Language Models
Jaideep Ray
Production LLM systems increasingly require machine-readable outputs: JSON objects, typed traces, regex-constrained fields, and tool-call schemas. This paper targets on-device and…