Posters

SMART: a novel self-supervised model tackles missing EHR data, improving patient health status prediction via missing-aware attention and robust pre-training.

SMART: a scalable, real-time multi-agent driving simulator using next-token prediction, achieves state-of-the-art results and zero-shot generalization.

SMALLTOLARGE (S2L) revolutionizes large language model (LLM) fine-tuning by using a small model to summarize training loss trajectories, enabling efficient data selection for larger models.

Stochastic gradient bandit algorithms now guaranteed to globally converge, using ANY constant learning rate!

SmMIL enhances medical image classification by introducing a novel smooth operator to improve instance-level localization in multiple instance learning, achieving state-of-the-art results.

SLTrain: Sparsity+low-rank pretraining boosts LLM efficiency by up to 73% memory reduction without performance loss!

SlowFocus significantly improves fine-grained temporal understanding in video LLMs by using mixed-frequency sampling and a novel multi-frequency attention mechanism.

SLowcal-SGD, a new local update method for distributed learning, provably outperforms Minibatch-SGD and Local-SGD in heterogeneous settings by using a slow querying technique, mitigating bias from loc…

Slot-VLM generates semantically decomposed video tokens using an Object-Event Slots module, improving video-language model performance.

SlotSSMs: a novel framework for modular sequence modeling, achieving significant performance gains by incorporating independent mechanisms and sparse interactions into State Space Models.

SlimSAM achieves near original SAM performance using 0.1% of its training data by employing a novel alternate slimming framework and disturbed Taylor pruning, significantly advancing data-efficient mo…

SlimGPT: Achieve near-optimal LLM structured pruning via Batched Greedy Pruning and Incremental Pruning Ratio, improving efficiency without sacrificing accuracy.

SLIM: A novel audio deepfake detection model leverages style-linguistics mismatch for superior generalization and explainability.

Scala: One-shot training enables flexible ViT inference!

SleeperNets: A universal backdoor attack against RL agents, achieving 100% success rate across diverse environments while preserving benign performance.

Self Logits Evolution Decoding (SLED) boosts LLM factuality by up to 20% without extra data or fine-tuning!

A novel slack-free spiking neural network efficiently solves the Hypergraph Minimum Vertex Cover problem on neuromorphic hardware, outperforming CPU-based methods in both speed and energy consumption.

SkipPredict optimizes scheduling by prioritizing cheap predictions and using expensive ones only when necessary, achieving cost-effective performance.

Skill-aware Mutual Information optimization enhances RL agent generalization across diverse tasks by distinguishing context embeddings based on skills, leading to improved zero-shot performance and ro…

SkiLD, a novel unsupervised skill discovery method, uses state factorization and a new objective function to learn skills inducing diverse interactions between state factors, outperforming existing me…