Search papers, labs, and topics across Lattice.
100 papers published across 3 labs.
RA-VLA transforms robotic adaptation by seamlessly integrating context retrieval with execution, achieving superior performance without the need for training.
RetrievalRouter redefines document retrieval efficiency by dynamically adapting to query needs, achieving superior accuracy and speed simultaneously.
A hybrid usability model reveals critical factors that can predict M-commerce app ratings with surprising accuracy.
Fine-tuned models can outperform leading systems in automated fact-checking, but their effectiveness is highly dependent on the domain and evaluation metrics used.
P4-DT outperformed human surrogates in predicting patient preferences, achieving an impressive 81.7% accuracy by leveraging contextual dilemmas.
RetrievalRouter redefines document retrieval efficiency by dynamically adapting to query needs, achieving superior accuracy and speed simultaneously.
A hybrid usability model reveals critical factors that can predict M-commerce app ratings with surprising accuracy.
Fine-tuned models can outperform leading systems in automated fact-checking, but their effectiveness is highly dependent on the domain and evaluation metrics used.
P4-DT outperformed human surrogates in predicting patient preferences, achieving an impressive 81.7% accuracy by leveraging contextual dilemmas.
User preferences can shift dramatically over time, but DUMoE's innovative approach captures these changes with remarkable accuracy, outperforming traditional models.
A learned shortlist can make private dense retrieval both accurate and practical, achieving near-full-corpus quality with a fraction of the computational cost.
DCEO outperforms traditional GMV proxies by directly optimizing causal effects, leading to a significant boost in long-term user value in e-commerce search.
Compiling knowledge graphs into LoRA adapters allows for zero-cost context retrieval, but knowledge stored is not transferable through similarity, posing a challenge for effective querying.
Bridging cognitive islands with a novel framework, PonsRAG boosts long narrative reasoning accuracy by over 11% through innovative evidence integration.
Reusing reasoning from previous queries can boost multi-hop QA accuracy while cutting down on token usage, transforming how we approach graph-based retrieval systems.
Bridging the semantic gap between images and text can dramatically enhance multimodal RAG performance, as shown by our novel Context-Enhanced MMKG framework.
A novel Dual-Transformer architecture boosts multi-camera view recommendation performance, achieving a 56.60% Precision@0.5 and enabling personalized editing style adaptation with just 20% of the video for training.
Geographic biases in LLMs persist even with retrieval-augmented generation, revealing that larger models don’t necessarily solve the problem.
Calibrating procedural relations in skill retrieval can boost task performance by over 10 points while streamlining execution efficiency.
Evidence weighting in Homo-RAG boosts retrieval effectiveness, achieving near-perfect relevance scores while enhancing the quality of gene function predictions for non-model organisms.
Achieving 100% verdict accuracy on compliance checks reveals the potential of visual-first approaches to revolutionize civil engineering workflows.
ReliableRAG achieves a substantial boost in factual accuracy for multi-hop QA by effectively filtering out deceptive misinformation through a reliability-driven approach.
Unsupervised query correction can achieve superior performance by cleverly encoding phonetic and visual similarities, avoiding the pitfalls of intent drift.
Candidate answer-side concepts infiltrate 7.40% of LLM-generated queries, challenging the integrity of information retrieval evaluations.
Role-aware matching in image retrieval can boost performance by up to 23% without the need for task-specific training.
Disentangling modal information in multi-modal recommendation leads to significant performance boosts, with D3ER outperforming traditional methods by effectively optimizing specialized models.
RA-VLA transforms robotic adaptation by seamlessly integrating context retrieval with execution, achieving superior performance without the need for training.
Human-written album reviews can dramatically enhance music retrieval models, yielding substantial gains in performance on complex queries.
DocPC slashes indexing time by 7.7x while boosting retrieval accuracy, redefining how we approach document-level visual search.
RDQ reveals that a distance-aware metric can significantly enhance the evaluation of retrieval systems by prioritizing both the relevance and the order of retrieved items.
PUMA slashes memory usage by up to 16x while maintaining or enhancing retrieval performance in multimodal tasks.
Event Tokens can significantly enhance LLM recommendation systems by capturing rich contextual information, leading to superior performance on industrial benchmarks.
TransRetrieval boosts recommendation recall by over 19 points while slashing computational costs by 85%, proving that smarter aggregation can outperform brute force scaling.
HSR reveals that modeling user preferences as a second-order dynamical system can dramatically enhance recommendation accuracy by capturing complex behavioral patterns often overlooked by traditional methods.
Users may prefer Q&A interfaces for their perceived ease, but they actually hinder knowledge construction compared to traditional document navigation.
MOTIF achieves a significant boost in cold-start multimodal recommendation performance by inferring user motivations and reconstructing item relationships, outperforming state-of-the-art methods.
Query expansions can backfire if not integrated properly, but AnchorQE shows how to harness them for up to 12.89% better retrieval performance.
Instantaneously adapting recommendation models to user requests without retraining could revolutionize how we personalize user experiences in real-time.
AdaptiveEmbed reveals that tailoring representation capacity to individual sample needs can dramatically enhance multimodal retrieval performance.
Achieving sublinear-regret in adversarial bandit submodular maximization under matroid constraints could redefine optimal strategies in complex decision-making scenarios.
PRQ-KMeans achieves up to 11.8% improvement in recommendation performance by rethinking how semantic identifiers are tokenized.
Cold-start recommendations can be effectively handled without retraining, achieving superior performance with a dual-encoder approach that keeps the index open to new items.
Migrating from gradient-boosted trees to deep learning not only preserves but enhances recommendation quality in dynamic customer support environments.
HSR boosts robot manipulation success rates by over 21% by leveraging hierarchical skill retrieval, even with minimal task-specific data.
Traditional metrics miss critical behavioral differences in RAG systems, but a new Bayesian framework reveals the true dynamics of retrieval and generation interactions.
MetaRAG achieves a superior accuracy-efficiency trade-off in agentic RAG by aligning decision-making with the model's internal beliefs, outperforming traditional RL methods.
Evidence Blindness can be mitigated by a novel navigation framework that organizes the corpus, leading to faster and more accurate evidence retrieval.
Personalized diet recommendations can now promote sustainability without sacrificing individual preferences, thanks to a novel constraint-aware decision-making model.
HMGCLIP achieves superior attribute discrimination in e-commerce by leveraging a novel hypergraph structure that captures both fine and coarse-grained product semantics.
Candidate admission can lead to significant rank loss for historical answers, but a new regularization technique effectively mitigates this issue without altering the embedding architecture.
Users can now find urban outdoor places that resonate with their activities and emotions, not just their categories, thanks to a novel intent-aware retrieval framework.
TAGR's innovative approach to real-time user intent modeling and ad tokenization leads to significant revenue gains in live-stream advertising.
CodeHID achieves a paradigm shift in code retrieval by organizing semantically similar snippets into a meaningful hierarchical index, leading to superior retrieval performance.
Mining high-quality samples from CTR data can dramatically enhance multimodal representation learning, leading to superior prediction accuracy.
A small, trainable module can dramatically improve localization accuracy in surgical video analysis, achieving the highest STG score on MedVidBench without retraining existing models.
KDPH's innovative use of Kent distributions allows it to absorb gradient conflicts, leading to superior performance in cross-modal hashing tasks.
The 9B WeMM-Embedding variant sets a new benchmark in multimodal embeddings, outperforming larger models while being deployed at scale across WeChat's diverse applications.
Synthetic question generation from knowledge graphs boosts retrieval precision and reasoning performance, even in the absence of labeled data.
Escalating computational effort only for the hardest cases boosts product linking accuracy from 68% to 77% while slashing costs by nearly 85%.
Evidence-dense top-ranked candidates can significantly boost recall while using fewer reranked documents, challenging conventional wisdom in RAG efficiency.
MoPLEx achieves up to 43.7% improvement in clustering accuracy by effectively learning from complex multi-way rankings, revealing the power of leveraging language models for preference optimization.
Retrieval accuracy alone doesn't guarantee that AI analysts will integrate crucial financial disclosures into their investment decisions, revealing a critical flaw in current evaluation methods.
RAGSentinel can filter out adversarial documents with high precision, ensuring that retrieval-augmented generation systems remain robust against sophisticated attacks.
CVE-SAI achieves the highest certified admission coverage while minimizing unsafe auto-induced exposure, revolutionizing how e-commerce attributes are indexed and verified.
A probabilistic framework reveals that synchronization overhead can negate the expected advantages of full-tree parallelism in distributed search systems.
SWIM redefines list evaluation by modeling user engagement as a survival process, leading to substantial gains in recommendation effectiveness.
Aligning semantic and collaborative signals in LLM-enhanced recommendations may actually hinder performance by ignoring valuable non-principal information.
Full-context interventions in Decision Transformers can dramatically shift predictions, but the effectiveness is highly dataset-dependent, revealing potential pitfalls in RTG conditioning.
Evidence-focused reasoning in RecGPT-Mobile-V2 boosts query quality by over 5% while slashing hard-failure rates to a mere 1.6%.
Crase achieves 3× higher recall at a third of the cost compared to existing deep research agents, redefining efficiency in scholarly search.
EviGraph's innovative approach to evidence construction boosts accuracy by over 30% compared to traditional methods, reshaping how agents validate information.
Tlow's innovative approach to item tokenization improves recommendation performance by over 10% in user engagement, even for cold-start items.
scrydb achieves a powerful blend of lexical and semantic search in SQLite, offering a lightweight yet effective solution for modern information retrieval challenges.
RAG systems are not just enhanced by external knowledge; they are also vulnerable to targeted attacks that can compromise accuracy, privacy, and fairness.
Trust evaluations can now adaptively integrate multi-source evidence while quantifying uncertainty, leading to more reliable collaborator selection in distributed systems.
Catastrophic performance drops occur when using unconstrained LLM-guided artist injection in music recommendation systems, emphasizing the need for careful model tuning.
BOAR reveals that addressing missing and unreliable auxiliary signals can lead to significant improvements in recommendation accuracy, especially for unseen items.
Stochastic separability of embedding manifolds reveals that distinct object categories can be linearly separated in high-dimensional spaces, provided certain conditions are met.
Historical relevance in time series forecasting is not universal; it’s structured and domain-dependent, revealing that future supervision can dramatically enhance predictive performance.
Zero-shot generalization in environmental classification is now achievable, with a hybrid model outperforming traditional methods by effectively leveraging retrieval techniques.
SA-RSQ achieves a remarkable balance between compact storage and high-quality representation, leading to significant performance boosts in real-world recommender systems.
Coarse indexing can dramatically speed up long-video retrieval without sacrificing answer quality, achieving up to 1.7x acceleration.
Organizing embeddings by activation patterns rather than treating them as fixed points can dramatically enhance retrieval efficiency and coverage in dense vector search.
EvoWiki achieves a remarkable 9.72% improvement in accuracy over existing methods, revolutionizing how we handle knowledge evolution in collaborative environments.
Web-enabled LLMs can boost corporate-event record accuracy by up to 48 percentage points, making them indispensable for financial institutions seeking reliable data.
Multimodal RAG systems can significantly improve retrieval quality, but their computational demands may not justify the gains in all scenarios.
Trigger items can dramatically enhance recommendation relevance, and CRRN leverages this to outperform existing methods in CTR prediction.
Negative transfer in cross-domain recommendations can be effectively mitigated with a dual-expert LLM approach that enhances item-level understanding.
ActPair achieves state-of-the-art performance in person anomaly detection by integrating action-aligned retrieval with efficient pairwise multimodal reranking, redefining how we approach context-dependent behavior analysis.
ASR errors can dramatically worsen the performance of advanced retrieval-augmented generation systems, with error amplification reaching up to 67%.
Traditional relevance proxies fail on hard negatives, but a new causal measurement approach reveals how to allocate context for up to 20.5% better recall in generative search.
Answer consistency in retrieval-augmented QA systems can deteriorate significantly, even when aggregate accuracy appears stable.
Memory retrieval accuracy can soar to 95.7% without any lexical matching, challenging traditional assumptions about language model dependencies on text.
Achieving 25% higher throughput than dense models, Giga-Embeddings redefines efficiency in high-quality text embedding generation.
Retrieval-augmented language models can significantly narrow the lexical frequency gap in syntactic contrast sensitivity, but they still leave room for improvement.
Restricting LoRA updates to a compact mid-network region can recover over half of the performance gains, challenging the notion that full-layer updates are necessary for effective fine-tuning.
MDA models outperform MDD models in dynamic collections, challenging assumptions about document dependence in IR systems.
Later entrants in the Shopify app ecosystem are outpacing early movers, revealing that timing alone does not guarantee success in competitive markets.
A novel interactive agent outperforms traditional LLMs by transforming vague candidate requests into precise sourcing outcomes, achieving unmatched relevance and coverage.
FashionEcoKG transforms fashion question answering by integrating a comprehensive knowledge graph, leading to a significant boost in retrieval and answer accuracy.
AdaWidth achieves state-of-the-art retrieval performance while drastically reducing the number of evaluated dimensions, adapting to the needs of each query.
Integrating item-based collaborative filtering into discrete diffusion models leads to a substantial boost in recommendation accuracy, outperforming existing methods.
Shifting from fixed-length tokens to semantic subwords can dramatically enhance the efficiency of attention mechanisms in generative recommenders.
HEGM not only stabilizes watch-time predictions but also boosts user engagement in real-world applications, outperforming its predecessor in both accuracy and interpretability.
Transforming Composed Image Retrieval into a training-free framework could redefine how we approach multimodal search tasks.
