Posters

Larger language models better predict brain activity in fMRI studies, with left-hemisphere prediction significantly increasing as model complexity scales up, reconciling classic aphasia findings with …

FM-Delta: Lossless compression halves cloud storage for massive fine-tuned language models, saving costs without sacrificing accuracy.

FlowTurbo: Blazing-fast, high-quality flow-based image generation via a velocity refiner!

FlowLLM revolutionizes material design by cleverly merging large language models and Riemannian flow matching, yielding a 300% boost in stable material generation!

Deep neural networks finally match human biological motion perception capabilities by leveraging patch-level optical flows and innovative neuron designs, achieving a 29% accuracy improvement.

ICTM efficiently solves linear inverse problems using flow priors by iteratively optimizing local MAP objectives, outperforming other flow-based methods.

FLORA enables efficient & private federated fine-tuning of LLMs via novel stacking-based heterogeneous low-rank adaptation, surpassing existing methods.

FlexSBDD, a novel deep generative model, accurately predicts flexible protein-ligand complex structures, generating high-affinity drug molecules while overcoming the limitations of rigid protein model…

FlexPlanner: Deep reinforcement learning solves flexible 3D floorplanning, improving wirelength and alignment significantly.

Brain’s flexible mapping of abstract domains is achieved via self-supervised extraction and projection of generalized velocity signals by grid cells, enabling efficient map generation.

Biologically-inspired DCnet neural network flexibly modulates visual processing based on context, outperforming existing models on visual search and attention tasks.

FlexCap generates controllable, region-specific image descriptions of varying lengths, achieving state-of-the-art zero-shot visual question answering.

A-SAM: Turning SAM’s inherent ambiguity into an advantage for controllable, diverse, and convincing ambiguous object segmentation.

Flatten Anything Model (FAM) revolutionizes neural surface parameterization with unsupervised learning, handling complex topologies and unstructured data fully automatically.

FLAME: A novel alignment method enhances large language model factuality by addressing hallucination in supervised fine-tuning and reinforcement learning, resulting in more accurate and helpful AI ass…

Bayesian best-arm identification algorithm achieves near-optimal sample complexity by incorporating an early-stopping criterion.

Two novel first-order algorithms, FOSL and MemCS, efficiently solve multi-block minimax bilevel optimization problems, significantly improving performance in deep AUC maximization and robust meta-lear…

First-order methods conquer linearly constrained bilevel optimization, achieving near-optimal convergence rates and enhancing high-dimensional applicability.

Meta-RL agents often fail to explore effectively in environments where optimal behavior requires sacrificing immediate rewards for greater future gains. First-Explore, a novel method, tackles this by…

FineStyle enables fine-grained controllable style personalization for text-to-image models using a novel concept-oriented data scaling and parameter-efficient adapter tuning, mitigating content leakag…