Posters
2024
Quantum Algorithms for Non-smooth Non-convex Optimization
·360 words·2 mins·
loading
·
loading
AI Theory
Optimization
🏢 Chinese University of Hong Kong
Quantum algorithms achieve speedups in non-smooth, non-convex optimization, outperforming classical methods by a factor of ε⁻²/³ in query complexity for finding (δ,ε)-Goldstein stationary points.
Quantum algorithm for large-scale market equilibrium computation
·643 words·4 mins·
loading
·
loading
AI Generated
AI Theory
Optimization
🏢 Centre for Quantum Technologies, National University of Singapore
Quantum speedup achieved for large-scale market equilibrium computation!
Quantitative Convergences of Lie Group Momentum Optimizers
·1602 words·8 mins·
loading
·
loading
Machine Learning
Optimization
🏢 Georgia Institute of Technology
Accelerated Lie group optimization achieved via a novel momentum algorithm (Lie NAG-SC) with proven convergence rates, surpassing existing methods in efficiency.
Quantifying the Gain in Weak-to-Strong Generalization
·2368 words·12 mins·
loading
·
loading
AI Generated
Natural Language Processing
Large Language Models
🏢 Stanford University
Weakly supervised strong models outperform weak models; this gain is precisely quantified by the strong model’s misfit error on weak labels.
Quantifying and Optimizing Global Faithfulness in Persona-driven Role-playing
·2570 words·13 mins·
loading
·
loading
Natural Language Processing
Dialogue Systems
🏢 UC San Diego
New APC metric precisely quantifies & optimizes global faithfulness in persona-driven role-playing, offering a fine-grained, explainable evaluation and improving AI character consistency.
Quantifying Aleatoric Uncertainty of the Treatment Effect: A Novel Orthogonal Learner
·2359 words·12 mins·
loading
·
loading
AI Theory
Causality
🏢 LMU Munich
New orthogonal learner quantifies treatment effect’s randomness, providing sharper insights beyond average effects.
QuanTA: Efficient High-Rank Fine-Tuning of LLMs with Quantum-Informed Tensor Adaptation
·3333 words·16 mins·
loading
·
loading
AI Generated
Natural Language Processing
Large Language Models
🏢 MIT
QuanTA: Quantum-inspired Tensor Adaptation efficiently fine-tunes LLMs with high-rank updates, surpassing low-rank methods like LoRA for complex tasks while minimizing additional parameters.
Quality-Improved and Property-Preserved Polarimetric Imaging via Complementarily Fusing
·1809 words·9 mins·
loading
·
loading
Computer Vision
Image Enhancement
🏢 Peking University
This paper introduces a novel three-phase neural network framework that significantly enhances the quality of polarimetric images by complementarily fusing degraded noisy and blurry snapshots, preserv…
Qualitative Mechanism Independence
·1560 words·8 mins·
loading
·
loading
AI Theory
Causality
🏢 Cornell University
Researchers introduce QIM-compatibility, a novel framework for modeling qualitative relationships in probability distributions using directed hypergraphs, significantly expanding beyond standard condi…
Quadratic Quantum Variational Monte Carlo
·1669 words·8 mins·
loading
·
loading
AI Theory
Optimization
🏢 University of Texas at Austin
Q2VMC, a novel quantum chemistry algorithm, drastically boosts the efficiency and accuracy of solving the Schrödinger equation using a quadratic update mechanism and neural network ansatzes.
QuadMamba: Learning Quadtree-based Selective Scan for Visual State Space Model
·2714 words·13 mins·
loading
·
loading
AI Generated
Computer Vision
Image Classification
🏢 Shanghai Jiao Tong University
QuadMamba: A novel vision model leveraging quadtree-based scanning for superior performance in visual tasks, achieving state-of-the-art results with linear-time complexity.
QT-ViT: Improving Linear Attention in ViT with Quadratic Taylor Expansion
·1611 words·8 mins·
loading
·
loading
AI Generated
Computer Vision
Image Classification
🏢 Advanced Micro Devices, Inc.
QT-ViT boosts Vision Transformer efficiency by using quadratic Taylor expansion to approximate self-attention, achieving state-of-the-art accuracy and speed.
QGFN: Controllable Greediness with Action Values
·3928 words·19 mins·
loading
·
loading
Machine Learning
Reinforcement Learning
🏢 Hong Kong University of Science and Technology
QGFN boosts Generative Flow Networks (GFNs) by cleverly combining their sampling policy with an action-value estimate, creating controllable and efficient generation of high-reward samples.
QBB: Quantization with Binary Bases for LLMs
·1816 words·9 mins·
loading
·
loading
Natural Language Processing
Large Language Models
🏢 Samsung AI Cambridge
QBB: A novel post-training quantization method for LLMs dramatically improves efficiency by replacing multiplications with summations, achieving state-of-the-art results with minimal accuracy loss.
Q-VLM: Post-training Quantization for Large Vision-Language Models
·2070 words·10 mins·
loading
·
loading
Multimodal Learning
Vision-Language Models
🏢 Tsinghua University
Q-VLM: A novel post-training quantization framework significantly compresses large vision-language models, boosting inference speed without sacrificing accuracy.
Q-Distribution guided Q-learning for offline reinforcement learning: Uncertainty penalized Q-value via consistency model
·4297 words·21 mins·
loading
·
loading
AI Generated
Machine Learning
Reinforcement Learning
🏢 Hong Kong University of Science and Technology
Offline RL struggles with OOD action overestimation. QDQ tackles this by penalizing uncertain Q-values using a consistency model, enhancing offline RL performance.
Putting Gale & Shapley to Work: Guaranteeing Stability Through Learning
·1809 words·9 mins·
loading
·
loading
AI Theory
Optimization
🏢 Penn State University
Researchers improve two-sided matching market algorithms by prioritizing stability through novel bandit-learning algorithms, providing theoretical bounds on sample complexity and demonstrating intrigu…
PureGen: Universal Data Purification for Train-Time Poison Defense via Generative Model Dynamics
·3344 words·16 mins·
loading
·
loading
Machine Learning
Deep Learning
🏢 UC Los Angeles
PUREGEN uses generative model dynamics to purify poisoned training data, providing a universal, effective, and efficient train-time defense against various data poisoning attacks.
PURE: Prompt Evolution with Graph ODE for Out-of-distribution Fluid Dynamics Modeling
·2009 words·10 mins·
loading
·
loading
Machine Learning
Deep Learning
🏢 Tencent
PURE: A novel method uses Graph ODE to adapt spatio-temporal forecasting models to various fluid dynamics scenarios, improving model adaptation to unseen parameters and long-term predictions.
Pure Message Passing Can Estimate Common Neighbor for Link Prediction
·2519 words·12 mins·
loading
·
loading
Machine Learning
Representation Learning
🏢 Computer Science and Engineering, University of Notre Dame
Pure message passing in graph neural networks can accurately estimate common neighbor heuristics for superior link prediction.