Search papers, labs, and topics across Lattice.
11
0
12
12
Extensive offline evaluation, human calibration, and online A/B tests show that SARA-7B generates more specific, polarity-consistent, and grounded rationales than strong MLLM baselines, while SARA-Ranker improves engagement and reduces negative feedback in production.
This work proposes Vid-PRE (Video Prompt Reasoner and Enhancer), a model-agnostic prompt rewriter that offloads the cognitive burden of reasoning to a dedicated VLM, and introduces VWG-Bench, a comprehensive benchmark spanning 9 reasoning dimensions and 38 fine-grained tasks.
Giving agent harness optimizers unconstrained edit scope reliably degrades performance; enforcing graduated, periodic permission scheduling ($1\to2\to3$) instead yields an immediate 20% jump in held-out task completion.
SWIM redefines list evaluation by modeling user engagement as a survival process, leading to substantial gains in recommendation effectiveness.
WebSwarm's innovative recursive delegation allows agents to not only search but also adaptively collaborate, leading to superior performance in complex web search tasks.
Retention models can now harness the power of post-conversion content without risking feature leakage, leading to more accurate predictions of user engagement.
Reranking in recommender systems can be revolutionized by shifting from local indices to generating global identifiers, enhancing robustness and user satisfaction.
UniMixer achieves state-of-the-art scaling in recommendation systems by unifying disparate architectures into a single framework that learns optimal token mixing patterns.
FlashEvaluator slashes the computational cost of evaluating multiple sequences in Generator-Evaluator frameworks while boosting accuracy by enabling direct cross-sequence comparisons.
SVD-Attention slashes the quadratic cost of attention to linear for recommendation tasks by exploiting the inherent low-rank structure of user behavior sequences, without sacrificing softmax.
Achieve lossless acceleration of ranking models by structurally re-parameterizing feature fusion matrix multiplication, sidestepping the accuracy drop common in lightweighting and distillation.
