23 citations · 46 across the 12 of their papers we have counts for
13 papers
Cosmos 3: Omnimodal World Models for Physical AI
NVIDIA, :, Aditi +293
We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-t…
Value-Aware Stochastic KV Cache Eviction for Reasoning Models
Ting-Yun Chang, Harvey Yiyun Fu, Deqing Fu +3
Reasoning models improve accuracy through extended chains of thought, but their long outputs create a memory and compute bottleneck. KV cache eviction methods reduce this cost by e…
SWE-IF: Aligning Code Evaluation with Human Preference
Ming Zhong, Xiang Zhou, Ting-Yun Chang +9
Large Language Models (LLMs) have catalyzed vibe coding, where users leverage LLMs to generate and iteratively refine code through natural language interactions until it passes the…
Why Do Some Inputs Break Low-Bit LLM Quantization?
Ting-Yun Chang, Muru Zhang, Jesse Thomason +1
Low-bit weight-only quantization significantly reduces the memory footprint of large language models (LLMs), but disproportionately affects certain examples. We analyze diverse 3-4…
When Parts Are Greater Than Sums: Individual LLM Components Can Outperform Full Models
Ting-Yun Chang, Jesse Thomason, Robin Jia
This paper studies in-context learning by decomposing the output of large language models into the individual contributions of attention heads and MLPs (components). We observe cur…
Do Localization Methods Actually Localize Memorized Data in LLMs? A Tale of Two Benchmarks
Ting-Yun Chang, Jesse Thomason, Robin Jia
The concept of localization in LLMs is often mentioned in prior work; however, methods for localization have never been systematically and directly evaluated. We propose two comple…
Data Curation Alone Can Stabilize In-context Learning
Ting-Yun Chang, Robin Jia
In-context learning (ICL) enables large language models (LLMs) to perform new tasks by prompting them with a sequence of training examples. However, it is known that ICL is very se…
CLiMB: A Continual Learning Benchmark for Vision-and-Language Tasks
Tejas Srinivasan, Ting-Yun Chang, Leticia Leonor Pinto Alva +3
Current state-of-the-art vision-and-language models are evaluated on tasks either individually or in a multi-task setting, overlooking the challenges of continually learning (CL) t…
Rethinking Why Intermediate-Task Fine-Tuning Works
Ting-Yun Chang, Chi-Jen Lu
Supplementary Training on Intermediate Labeled-data Tasks (STILTs) is a widely applied technique, which first fine-tunes the pretrained language models on an intermediate task befo…
Go Beyond Plain Fine-tuning: Improving Pretrained Models for Social Commonsense
Ting-Yun Chang, Yang Liu, Karthik Gopalakrishnan +3
Pretrained language models have demonstrated outstanding performance in many NLP tasks recently. However, their social intelligence, which requires commonsense reasoning about the…
Incorporating Commonsense Knowledge Graph in Pretrained Models for Social Commonsense Tasks
Ting-Yun Chang, Yang Liu, Karthik Gopalakrishnan +3
Pretrained language models have excelled at many NLP tasks recently; however, their social intelligence is still unsatisfactory. To enable this, machines need to have a more genera…
TinyGAN: Distilling BigGAN for Conditional Image Generation
Ting-Yun Chang, Chi-Jen Lu
Generative Adversarial Networks (GANs) have become a powerful approach for generative image modeling. However, GANs are notorious for their training instability, especially on larg…
xSense: Learning Sense-Separated Sparse Representations and Textual Definitions for Explainable Word Sense Networks
Ting-Yun Chang, Ta-Chung Chi, Shang-Chi Tsai +1
Despite the success achieved on various natural language processing tasks, word embeddings are difficult to interpret due to the dense vector representations. This paper focuses on…