Retrieval pipelines, vector search, chunking and reranking: how models are grounded in a corpus instead of their weights.
arXiv:2610. 09760v1 Announce Type: new Abstract: Transformers have in-context learning capabilities, where some known learning algorithms can be executed in the forward pass through the model.
Video anomaly detection (VAD) is a fundamental and safety-critical task in computer vision. Recent generative approaches detect anomalies from a distributional perspective, but remain limited by local...
Muon has motivated a growing family of optimizers that compose multiple matrix normalizations, but these methods remain fragmented and lack a unified perspective. We introduce chained linear minimizat...
Latent world models have shown a remarkable ability to predict future states and to plan in the real world. In practice, however, we lack a principled way to estimate how their capabilities scale with...
Prompted embedding models have recently received increasing attention, particularly for retrieval, where detailed retrieval instructions are provided as part of the retrieval prompt. Several new datas...
Interpretability offers a solution to safe autonomy by providing transparency into an agent's underlying decision-making model. Within sequential-decision making tasks, differentiable decision trees (...
Medical vision-language models (VLMs) have shown increasing potential for clinical image interpretation. However, these models still struggle to interpret multi-phenotype lesions whose diagnosis requi...
When does distribution alignment help a frozen foundation-model embedding generalize across acoustic domains? For cross-domain mosquito-species classification we report a strength-monotonic law: the s...
Entity Alignment (EA) identifies equivalent entities across knowledge graphs and is critical for knowledge base integration and ontology merging. Evaluating EA systems at scale requires expensive expe...
A retrieval-augmented model can match a document without relying on it. Controlled knowledge conflicts make source choice observable and let us ask a second question that prediction alone cannot answe...
Zero-shot Chinese character recognition (ZS-CCR) aims to recognize characters whose categories are never observed during training, and typically relies on the compositional structure shared between se...
Retrieval diversification is widely available in retrieval-augmented generation (RAG) frameworks, yet prior studies disagree on whether it improves retrieval and answer quality. We show that its effec...
CausalBind introduces a causal modeling framework for protein‑molecule virtual screening that focuses on sparse cross‑modality interactions rather than dense holistic alignment. By formalizing a V‑structure causal model with Heckman‑style selection, the authors prove that latent interaction concepts are identifiable only under sparsity constraints and propose three implementation variants. Experiments on DUD‑E and LIT‑PCBA benchmarks demonstrate that all variants outperform strong retrieval baselines, especially in early enrichment and out‑of‑distribution generalization.
STRUCTURALCOST is a self‑paced reading dataset comprising 475 participants and 40,800 observations that isolates the processing cost of long‑distance subject‑verb dependency resolution. The study replicates a known psycholinguistic finding at NLP scale: human reading times at the main verb increase with dependency length, driven by syntactic embedding beyond linear distance. Various language models—n‑gram, SSMs, and transformers—partially mirror this graded difficulty profile but consistently underestimate the integration cost humans incur, a gap that persists across architectures and model sizes.
The paper evaluates five multivariate two-sample drift detectors on large‑scale e‑commerce data, including a 137.5‑million‑row Trendyol feature table. Distributed Maximum Mean Discrepancy with Random Fourier Features on Apache Spark shows strong performance, achieving a Pearson correlation of 0.940 with expected drift magnitude, an 80.4% true positive rate, and a 3.2% false positive rate. In contrast, the per‑dimension Kolmogorov‑Smirnov test fails due to statistic saturation from ID‑like columns, and detectors struggle under weak drift configurations, underscoring limitations in current sampling designs and sensitivity analysis.
The paper introduces Multi-Probe Zero Collision Hash (MPZCH), an indexing mechanism that uses linear probing to reduce or eliminate collisions in embedding tables for large-scale recommendation systems. MPZCH employs auxiliary tensors and CUDA kernels to support configurable probing and active eviction, thereby retiring obsolete IDs and resetting reassigned slots to prevent stale embeddings. Experiments show that MPZCH achieves zero collisions for user embeddings, improves item embedding freshness and quality, and maintains training and inference performance comparable to existing methods.
The paper introduces SCALE (State-CAlibrated Latent Embeddings), a technique that aligns pairwise latent distances with task-relevant state-space distances in joint-embedding predictive world models. By adding SCALE to the LeWorldModel (LeWM) objective, the authors preserve LeWM’s architecture while ensuring that latent geometry reflects meaningful task outcomes. Experiments demonstrate that SCALE improves planning success across manipulation and navigation tasks with various solvers, and the authors analyze how the method reshapes representation geometry to support planning.
The paper presents a method for early triage of misinformation by forecasting the future growth of social media information cascades from the first 30 minutes of activity. Using features such as node count, structural depth entropy, and temporal arrival entropy, the combined model improves predictive accuracy on 352 test trees, especially for high‑activity cascades. Additionally, the study analyzes early reply patterns in PHEME threads and evaluates evidence retrieval in a Check‑COVID setting, proposing an integrated human‑review system that leverages these early signals for misinformation triage.
The paper introduces ELMRec, an enhanced LLM-based recommender that improves high-order interaction awareness by using whole-word embeddings to better interpret graph-constructed user-item interactions without graph pre-training. It also identifies that LLMs tend to prioritize earlier user interactions over recent ones and proposes a reranking solution. ELMRec achieves superior performance compared to state-of-the-art methods in both direct and sequential recommendation tasks.
The paper identifies a representational bottleneck in 3D CT embeddings that limits pathology coverage in automated radiology reports. It introduces AdaRAG-CT, an adaptive retrieval-augmented framework that supplements visual features with controlled textual retrieval to improve report generation. On the CT-RATE benchmark, AdaRAG-CT achieves state‑of‑the‑art clinical efficacy, raising Clinical F1 from 0.420 to 0.480.