Jin Song Dong
12 papers · 2022–2026 · 6 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
π Cross-Pollinator (15) π Conference Polyglot (5) π Interdisciplinary Bridge πΊοΈ Taxonomy Completionist (15) π§ Keyword Pioneer
π
Renaissance Researcher
(5)
π
Grand Slam
π
Century Club
(10)
π
Trend Setter
Conferences
AAAI (3)
ICML (3)
ACL (2)
NIPS (2)
ICLR (1)
IJCAI (1)
Top co-authors
Keywords
large language model
(3)
representation learning
(1)
label noise detection
(1)
few-shot learning
(1)
embedding space
(1)
knowledge distillation
(1)
code generation
(1)
deep learning
(1)
formal methods
(1)
model debugging
(1)
gradient descent
(1)
generalization error
(1)
first-order logic
(1)
influence function
(1)
deep metric learning
(1)
noisy label
(1)
harmful content detection
(1)
attention kernel
(1)
cuda kernel
(1)
intermediate representation
(1)
Papers
CuBridge: An LLM-Based Framework for Understanding and Reconstructing High-Performance Attention Kernels
ACL 2026
Inverting the Shield: Systematically Generating Safety Tests from Policy Specifications
ACL 2026
Failures to Surface Harmful Contents in Video Large Language Models
AAAI 2026
Few-Shot Precise Event Spotting via Unified Multi-Entity Graph and Distillation
AAAI 2026
Position: Trustworthy AI Agents Require the Integration of Large Language Models and Formal Methods
ICML 2025
$F^3Set$: Towards Analyzing Fast, Frequent, and Fine-grained Events from Videos
ICLR 2025
Clustering Properties of Self-Supervised Learning
ICML 2025
Defending LVLMs Against Vision Attacks Through Partial-Perception Supervision
ICML 2025
On-the-Fly Adapting Code Summarization on Trainable Cost-Effective Language Models
NIPS 2023
Temporality Spatialization: A Scalable and Faithful Time-Travelling Visualization for Deep Classifier Training
IJCAI 2022
Debugging and Explaining Metric Learning Approaches: An Influence Function Based Perspective
NIPS 2022
DeepVisualInsight: Time-Travelling Visualization for Spatio-Temporal Causality of Deep Classification Training
AAAI 2022