🏗️ Building on HF

Kshitij Thakkar PRO

kshitijthakkar

·

AI & ML interests

Building the evaluation and observability layer for AI. Creator of TraceVerse—turning real-world LLM interactions into datasets, benchmarks, and cost-efficient model insights.

Organizations

kshitijthakkar 's Spaces 41

GuardianTails

Pet Health Intelligence Platform

2604.20098

Explore and collaborate on your project logbook

Repro - Finite and Corruption-Robust Regret Bounds in Online Inverse Linear Optimization under M-Convex Action Sets

Explore and collaborate on a research logbook online

Repro - TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning

Explore experiment logs and sync findings with a coding agent

Repro - Gaussian Mean Field Variational Inference can Overestimate Predictive Variance

Explore research logbook and sync with your coding agent

Reproducing GRPO's Loss, Dynamics, and Success Amplification (arXiv:2503.06639)

Browse a research logbook and collaborate with an AI agent

Repro - Dependence-Aware Label Aggregation via Ising Models

Explore and edit experiment logbooks with AI agent help

Repro - Do-Calculus Derivation Graphs

Explore do‑calculus derivation graphs and sync with an AI agent

DynaSchedBench reproduction

View and sync experiment logs with an AI agent

Repro - Judging What We Cannot Solve: Consequence-Based Utility for Oracle-Free Evaluation

Explore logs and collaborate with an AI coding agent

Repro - Provable Benefits of RLVR over SFT: Learning to Backtrack Efficiently

Collaborate on a logbook with an AI coding agent

ICML Provincia

Replay the ICML 2026 Agent‑Repro challenge logs

Repro - Row-Stochastic vs Doubly-Stochastic Decentralized Learning

Explore and sync experiment logs with your coding agent

Repro - Unsupervised Disentanglement Without Compromises

Browse and collaborate on experiment logbooks

Reproducing ITCR: inference-time conformal factuality control

Browse a research logbook and collaborate with an AI agent

Repro - BrokenMath: Sycophancy in Theorem Proving

Explore and manage research logbook entries online

Repro - Dimension-Free KL Bounds for Underdamped Langevin MC

Explore project logbook and collaborate with an AI agent

Reproducing: Success Conditioning as Policy Improvement (arXiv:2601.18175)

Explore research logbook and sync with your AI agent

Repro - Learning Treatment Allocations with Risk Control Under Partial Identifiability

Collaborate on a research logbook with an AI agent

Repro - When Actions Go Off-Task: DeAction and MisActBench

Explore action logs and sync findings with your coding agent

Repro - Asymptotically Optimal Sequential Testing with Markovian Data

Explore and sync experiment logs with an AI coding agent

Repro - Envy-Free Allocation of Indivisible Goods via Noisy Queries

Browse experiment logs and sync findings with your coding agent

Repro - How to Correctly Report LLM-as-a-Judge Evaluations

Log and share LLM evaluation findings in a collaborative notebook

Repro - A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness

Browse and collaborate on a research logbook with AI assistance