University of Texas at Austin

university

Verified

https://www.utexas.edu

AI & ML interests

None defined yet.

Recent Activity

cranial-xix submitted a paper 23 days ago

Maglev: Sliding Recurrent Memory

FOGmiaow submitted a paper about 2 months ago

ISO: An RLVR-Native Optimization Stack

pbansal submitted a paper about 2 months ago

Token Time Continuous Diffusion for Language Modeling

View all activity

Papers

Maglev: Sliding Recurrent Memory

ISO: An RLVR-Native Optimization Stack

View all Papers

submitted a paper to Daily Papers 23 days ago

Maglev: Sliding Recurrent Memory

Paper • 2608.02870 • Published Aug 5 • 14

submitted a paper to Daily Papers about 2 months ago

ISO: An RLVR-Native Optimization Stack

Paper • 2607.19331 • Published Jul 21 • 9

submitted a paper to Daily Papers about 2 months ago

Token Time Continuous Diffusion for Language Modeling

Paper • 2607.14106 • Published May 7 • 9

updated 2 datasets 2 months ago

UTEXAS/glioma-cell-radiation

Viewer • Updated Jun 27 • 49.8k • 1.12k

UTEXAS/glioma-cell-chemoradiation

Viewer • Updated Jun 27 • 20.5k • 378

published a dataset 2 months ago

UTEXAS/glioma-cell-chemoradiation

Viewer • Updated Jun 27 • 20.5k • 378

updated a dataset 2 months ago

UTEXAS/glioma-cell-chemo

Viewer • Updated Jun 27 • 69.2k • 381

submitted a paper to Daily Papers 3 months ago

CalVerT: Augmenting Agents with Calibrated Verifier Telemetry Improves Action and Learning in Knowledge-Intensive Tasks

Paper • 2606.21777 • Published Jun 19 • 5

submitted a paper to Daily Papers 3 months ago

No Hidden Prompts Needed! You Can Game AI Peer Review with Presentation-Only Revisions

Paper • 2606.13044 • Published Jun 11 • 11

published a dataset 3 months ago

UTEXAS/glioma-cell-chemo

Viewer • Updated Jun 27 • 69.2k • 381

published a dataset 4 months ago

UTEXAS/glioma-cell-radiation

Viewer • Updated Jun 27 • 49.8k • 1.12k

authored a paper 5 months ago

Discourse Diversity in Multi-Turn Empathic Dialogue

Paper • 2604.11742 • Published Apr 13 • 1

authored a paper 5 months ago

ParaSpeechCLAP: A Dual-Encoder Speech-Text Model for Rich Stylistic Language-Audio Pretraining

Paper • 2603.28737 • Published Mar 30

submitted a paper to Daily Papers 6 months ago

TERMINATOR: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought Reasoning

Paper • 2603.12529 • Published Mar 13 • 19

submitted 2 papers to Daily Papers 6 months ago

Beyond Test-Time Training: Learning to Reason via Hardware-Efficient Optimal Control

Paper • 2603.09221 • Published Mar 10 • 1

nabla-Reasoner: LLM Reasoning via Test-Time Gradient Descent in Latent Space

Paper • 2603.04948 • Published Mar 5 • 2

submitted a paper to Daily Papers 7 months ago

EntRGi: Entropy Aware Reward Guidance for Diffusion Language Models

Paper • 2602.05000 • Published Feb 4 • 2

posted an update 8 months ago

Post

1655

Are you familiar with reverse residual connections or looping in language models?

Excited to share my Looped-GPT blog post and codebase 🚀
https://github.com/sanyalsunny111/Looped-GPT

TL;DR: looping during pre-training improves generalization.

Plot shows GPT2 LMs pre-trained with 15.73B OWT tokens

P.S. This is my first post here — I have ~4 followers and zero expectations for reach 😄

3 replies

·

authored a paper 9 months ago

Mitigating Catastrophic Forgetting in Mathematical Reasoning Finetuning through Mixed Training

Paper • 2512.13706 • Published Dec 5, 2025 • 1

authored a paper 9 months ago

Cautious Weight Decay

Paper • 2510.12402 • Published Oct 14, 2025 • 10