WinDeskGround is a parametrically generated benchmark of 1,356 instruction-target pairs that reveals accuracy declines in state-of-the-art MLLMs under partial occlusion in multi-window GUI settings.
Title resolution pending
27 Pith papers cite this work. Polarity classification is still indexing.
27
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
FeDPM learns and aligns local discrete prototypical memories across domains to create a unified discrete latent space for LLM-based time series foundation models in a federated setting.
GhostServe applies erasure coding to KV cache in host memory for fast recovery from failures in LLM serving, cutting checkpointing latency up to 2.7x and recovery latency 2.1x versus prior methods.
Parallel thinking in LLMs suffers from overscaling where fixed global budgets waste samples; LanBo predicts per-sample budgets from latent states to raise utilization without hurting accuracy.
A renewal-reward analysis yields a closed-form mean-field rule for the optimal Attention/FFN provisioning ratio in disaggregated LLM serving that accounts for stochastic KV-cache growth and matches simulation optima within 10%.
Diagnoses mask prior drift and positional attention collapse in LDVLMs and introduces two plug-and-play decoding interventions that raise long-form generation quality without retraining.
Anon optimizer uses tunable adaptivity and incremental delay update to achieve convergence guarantees and outperform existing methods on image classification, diffusion, and language modeling tasks.
Extracting task vectors from the offline dataset for policy training improves zero-shot offline RL performance by an average of 20% over random sampling baselines.
Dataset-level metrics in diffusion language models mask substantial sample-level non-determinism that varies with model and system factors, which a new Factor Variance Attribution metric can decompose.
BiSpikCLM is the first fully binary spiking MatMul-free causal language model that matches ANN performance on generation tasks using only 4-6 percent of the compute via softmax-free spiking attention and spike-aware distillation.
iAmTime is a time-series foundation model that uses instruction-conditioned in-context learning from demonstrations to perform zero-shot adaptation on forecasting, imputation, classification, and related tasks.
SiameseNorm is a two-stream architecture that reconciles Pre-Norm and Post-Norm in Transformers by coupling streams via shared residual blocks, yielding performance gains with maintained stability on language, vision, and diffusion models.
SPOT projects prompts to a tau-safe set via total variation to cut inappropriate content 14-44% relative to baselines while preserving benign prompt behavior in frozen T2I models.
TGR performs manifold-informed latent foresight search to boost trajectory coverage in long-context reasoning tasks by up to 13 AUC points with minimal overhead.
BOOST delivers 1.46-2.27x end-to-end speedups for low-rank bottleneck LLMs by redesigning tensor parallelism around the bottleneck structure plus supporting optimizations.
AlphaCast is a training-free LLM framework that performs interactive multi-stage reasoning for time series forecasting by integrating feature extraction, knowledge bases, case libraries, and contextual pools.
MoE-LLaVA applies mixture-of-experts sparsity to LVLMs via MoE-Tuning, delivering LLaVA-1.5-7B level visual understanding and better hallucination resistance with only ~3B active parameters.
Properly filtered web data from CommonCrawl alone trains LLMs that significantly outperform models trained on The Pile, with 600 billion tokens and 1.3B/7.5B parameter models released.
PaLM-E is a single 562B-parameter multimodal model that performs embodied reasoning tasks like robotic manipulation planning and visual question answering by interleaving vision, state, and text inputs with positive transfer from joint training on language and robotics data.
FAAST performs test-time supervised adaptation by analytically deriving fast weights from examples in one forward pass, matching backprop performance with over 90% less adaptation time and up to 95% memory savings versus memory-based methods.
DRAFT decouples agent safety judgment into latent extraction and reasoning stages, raising average benchmark accuracy from 63.27% to 91.18%.
The paper analyzes CPU bottlenecks in agentic AI serving, selects representative workloads, and demonstrates that CPU-aware scheduling optimizations COMB and MAS can reduce P50 latency by up to 1.7x and total latency by up to 2.49x on two hardware systems.
Table-specific pretraining of Llama-2 yields significant gains on zero-shot, few-shot, and in-context tabular prediction tasks over prior benchmarks.
citing papers explorer
-
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
MoE-LLaVA applies mixture-of-experts sparsity to LVLMs via MoE-Tuning, delivering LLaVA-1.5-7B level visual understanding and better hallucination resistance with only ~3B active parameters.
-
Unlock the Potential of Large Language Models for Predictive Tabular Tasks in Data Science with Table-Specific Pretraining
Table-specific pretraining of Llama-2 yields significant gains on zero-shot, few-shot, and in-context tabular prediction tasks over prior benchmarks.