Search papers, labs, and topics across Lattice.
100 papers published across 8 labs.
Most observed IPv6 Fragment-EH patterns are simple within-packet co-occurrences, not complex temporal network rules, a finding rigorously established by a novel negative-control protocol.
Asynchronous learning communities don't just fragment; they predictably fragment *before* converging around deadlines, a pattern driven by participation shifts, not just behavioral alignment.
A novel contrastive feedback mechanism allows LLM agents to learn skill proposals that are aligned with the current actor's behavior without requiring expensive skill evaluations.
LLM agents can now reliably track and fulfill complex deliverable requirements across numerous files without losing context, thanks to an environment-side "tab" that acts as a persistent, shared task ledger.
Agnostic smoothed online learning is now achievable without knowing the base measure or smoothing parameter, matching statistical learning's flexibility with online learning's robustness.
A novel contrastive feedback mechanism allows LLM agents to learn skill proposals that are aligned with the current actor's behavior without requiring expensive skill evaluations.
LLM agents can now reliably track and fulfill complex deliverable requirements across numerous files without losing context, thanks to an environment-side "tab" that acts as a persistent, shared task ledger.
Agnostic smoothed online learning is now achievable without knowing the base measure or smoothing parameter, matching statistical learning's flexibility with online learning's robustness.
Fine-grained emotion control in TTS is now achievable at inference time with minimal speaker or content degradation, bypassing the need for expensive model retraining.
Uncoded crash factors like phone use and medical episodes are vastly undercounted in police records, with narratives revealing 2x more injury crashes than coded fields.
The optimal strategy for the cow-path problem is provably cyclic, meaning no search algorithm can outperform a fixed path traversal order.
Machine learning agents discover new exact values for notoriously difficult-to-compute link invariants like slice genus and unknotting number.
Small Language Models can parse ML pipeline structures, but their insights into data science practices diverge significantly from other classifiers, highlighting the critical impact of taxonomy definition.
Generative orthogonal learning unlocks rate double robustness and quasi-oracle efficiency for estimating distributional treatment effects under complex time-varying confounding.
Inertial activity recognition models can now learn rich, global motion structure from unlabeled data by discovering and grouping motion primitives, achieving state-of-the-art performance with orders of magnitude less data.
Spatially dependent data breaks standard conformal prediction; this method achieves exact coverage and oracle efficiency by sequentially whitening residuals, even with nonrandom sampling.
Provably private tabular classification is now achievable with a single forward pass, outperforming baselines and drastically cutting fitting time.
Sim racing drivers' braking and strategy behaviors are not car-interchangeable, challenging existing profiling assumptions.
Adversarial bandits with $m$-set choices can suffer drastically slower learning rates than previously thought, with regret scaling quadratically with $m$ in some settings.
Black-box text-to-music generators can be steered to precisely match complex user intent by an agent that iteratively refines prompts based on rubric-based audio verification.
English skeletons offer a minor, inconsistent boost in multilingual math reasoning, with optimal language choice proving highly context-dependent.
Key contribution not extracted.
Interactive video generation systems can now robustly handle latent dependencies by transforming unstructured inputs into executable constraints, achieving a 10.3-point average improvement.
A single, lightweight model-internal mechanism can effectively bridge the gap between corpus retrieval and long-context inference, outperforming specialized SOTA systems on both.
Personal LLM agents can significantly improve recommendation quality by mediating platform rankings with cross-platform history, but naive mediation introduces substantial harmful overrides.
A two-input Fredholm backstepping design achieves rapid stabilization of the Kuramoto--Sivashinsky equation, even with spatially varying anti-diffusion and unstable double eigenvalues, by rendering the plant controllable via one input before stabilizing with the other.
Jointly diffusing token and cluster embeddings unlocks significant gains in continuous diffusion language models, achieving state-of-the-art perplexity and reasoning performance with minimal overhead.
Misspecifying the coupling mechanism for ordinal data in GP-SCMs inflates counterfactual error threefold, even when observational fit is high, and this error plateaus with more data.
A single interpolation coefficient unlocks variance-optimal off-policy evaluation by seamlessly blending the stability of cluster-level weights with the unbiasedness of action-level importance sampling.
Context compaction in long-horizon agents is surprisingly insensitive to recent agent behavior, with simple "has-written" heuristics measuring trajectory phase rather than predictive power.
The geometric intuition behind SVD is revealed as the direct algorithmic engine driving PCA, kernel methods, and PageRank, unifying their proofs and implementations.
Training-free TFMs can outperform trained GCNs in calibration for node classification, and a novel homophily-gated approach significantly shrinks prediction sets without sacrificing validity.
Quantum circuits can unlock compositional generalization in multimodal AI, even on noisy hardware.
Heterogeneous load forecasting scales efficiently by learning a small bank of low-rank adaptations for a shared probabilistic model, outperforming independent models and monolithic Transformers.
Multimodal geometric scores are surprisingly sensitive to the *direction* of modality degradation, not just its magnitude, revealing a critical geometric dependency previously overlooked.
Boosted trees with rolling calibration forecast German redispatch volumes accurately under significant data delays, but nominal coverage hides critical underperformance during extreme congestion.
Latent signals for unanswerability generalize robustly across diverse question-answering datasets, but their utility in multi-turn dialogue hinges on capturing conversational dynamics rather than just static missing information.
Strict orthogonal parameter isolation in LLM continual learning paradoxically harms knowledge transfer; CoDe-LoRA breaks this dilemma by decoupling universal and task-specific knowledge.
Generative transport reveals a hidden time-reflection symmetry in interpolants, enabling exact, non-parametric two-sample testing without density estimation.
Discovering complex, multi-attribute logical dependencies in tabular data is now scalable, revealing non-deterministic relationships previously hidden by computational limits.
A leading clinical AI platform matches or exceeds commercial ambient scribes on note quality, demonstrating the power of configurable API-driven generation.
A small, causally identified set of attention heads reliably detects network infrastructure information across diverse LLMs, but their responsibility may reside in single neurons or be distributed across the head depending on the model.
Existing autoregressive post-training weights, often discarded after diffusion model conversion, can be directly applied to significantly boost dLLM performance, even outperforming dedicated diffusion post-training when combined.
Most observed IPv6 Fragment-EH patterns are simple within-packet co-occurrences, not complex temporal network rules, a finding rigorously established by a novel negative-control protocol.
COMET scores are demonstrably biased by target script, with script identity explaining over 20% of variance and degrading human agreement, revealing a fundamental flaw in cross-script MT evaluation.
Expert counseling quality assessment transfers surprisingly well across domains, with LLM-extracted instrument-based constructs outperforming in-domain training and even beating generic dialogue features.
LLMs can now generate novel research ideas by explicitly learning to synthesize literature through structured "anchor" specifications, outperforming prior prompting and feedback methods.
Amortized Bayesian causal inference for hidden confounding is now possible via a single transformer pass, achieving state-of-the-art CATE estimation without hyperparameter tuning.
Even frontier LLMs misattribute incidental patient conversation in over 3% of clinical notes, and background noise contaminates transcripts in nearly half of mock consultations.
AI ideas can be reliably detected even when the prose is human-written, outperforming traditional detectors that focus on writing style.
Decision models can achieve order-invariant option scoring without sacrificing accuracy by treating each option independently, bypassing the need for complex sequence modeling.
A transformer pretrained on synthetic, never-repeating "languages" learns to infer and predict real-world text and perform complex reasoning tasks, suggesting a fundamental mechanism for language acquisition.
Aligning DLM generation with reference distributions via feature-space MMD significantly boosts performance on perplexity and reasoning tasks without retraining the entire model.
Discover hierarchical cluster structure across multiple scales in directed graphs without knowing the number of clusters beforehand, even across sparse bridges.
A novel two-stream architecture disentangles temporal context from spatial refinement in EEG source imaging, achieving superior localization accuracy without compromising inference speed.
A novel cross-domain recommendation framework achieves state-of-the-art book-to-music mood matching by fusing sentiment embeddings with LLM-driven emotional quadrant classification.
Explicitly linking multimodal patient data to biomedical knowledge graphs unlocks significant gains for clinical LLMs, with performance plummeting when these crucial connections are removed.
A single pre-trained transformer now solves matrix completion across causal inference, recommendation, and tabular imputation tasks by treating matrices natively.
Simple vMF likelihood modeling unlocks robust label-free speaker embedding enhancement, outperforming complex diffusion baselines on challenging acoustic mismatches.
Local AI models now predict user well-being from passive computer activity, offering explainable feedback directly within a workplace sensing application.
Binary sources allow for strictly smaller optimal representations in the Information Bottleneck than previously thought.
Generative auctions can now achieve externality-aware whole-feed optimization with exact strategyproofness using a transformer, overcoming a critical deployment barrier.
Amortized spike-and-slab inference for linear models with interactions achieves state-of-the-art variable selection accuracy with a single forward pass, eliminating costly per-dataset MCMC.
Autoregressive latent priors learned directly from data, not fixed Gaussians, unlock VAEs for high-fidelity sequential generation with diffusion-level quality and faster inference.
TSFMs for predictive control fail to capture system dynamics without explicit control excitation in their inference context, even with low forecasting error.
Recover a full Pareto stationary front of multi-objective solutions with the computational cost of standard single-objective training.
Text generation evaluation can now directly measure distributional divergence using hidden states and double sliced Wasserstein distance, bypassing perplexity's limitations for non-autoregressive models.
FairProp embeds fairness directly into GNN message passing, achieving state-of-the-art fairness-utility trade-offs by jointly optimizing group means and within-group covariance.
Kernel ensembles with Survival Forests and Beran estimators significantly outperform single models for patient-specific survival prediction.
Medical LLMs can achieve superior QA performance by dynamically selecting specialized adaptation components based on question context and clinical tags, surpassing static parameter-efficient methods.
Influence functions, not just activation averages, unlock more thematic and robust concept steering in LLMs.
Personalized suicide risk forecasting is significantly improved by embedding patients in a latent space that models similarity, outperforming pooled and per-patient models by up to 30% for first-occurrence event prediction.
Clinical QA models can now achieve substantial gains by leveraging a composite reward that fuses external ontology concept overlap with an LLM judge, outperforming SFT and capturing nuanced factual correctness.
Abusive language detection can be reframed as sequence labeling to pinpoint abusive spans and targets, offering localized moderation insights beyond simple sentence classification.
Flowcharts break silently across aspect ratios, but a new agentic pipeline with VLM critics now guarantees connectivity and editability.
Prompt optimization methods that iteratively rewrite prompts with LLM feedback often overfit to training data, but a new regularization framework achieves up to +16.5% OOD generalization gains by controlling prompt representation.
LLMs forget user intent after a rejected suggestion, treating it as active even when superseded, leading to task failure.
Choice-based models like Jev can accurately identify graph adjacencies but fail to grasp broader structural correctness, even with joint graph-text inputs.
A 32B model in Marathi performs like a 10B model in English, revealing how model size and reasoning can bridge language resource gaps in cross-lingual transfer.
A single tokenizer can now slash token counts for non-English scripts by 20-40% without increasing English tokenization cost.
Unlocking latent topological structures with discrete Morse theory boosts graph diffusion model performance, offering theoretical stability guarantees.
Optimal $\mathcal{O}(\varepsilon^{-4-2/p})$ convergence for stochastic bilevel optimization is achieved with a single-loop first-order method, matching a new lower bound.
Robust optimal policies and values for RMDPs are now computable in strongly polynomial time via a single LP, unifying and extending prior results across diverse uncertainty models.
Malware analysis just got a new tool: S4 models unlock long-range dependency detection in execution flows, outperforming traditional deep learning for multi-class classification.
Standard GNNs and Transformers can now capture complex multi-way interactions by simply adding higher-order positional encodings derived from clique complexes, boosting performance on tasks requiring rich relational understanding.
Relational learning with tabular foundation models can now bypass feature explosion and interaction blindness by propagating information directly through the schema graph.
Per-user normalization of wrist EDA data can inflate affect recognition performance by up to 4x, masking the true benefit of pretraining and making published results incomparable.
LLM agents can now build lasting, reusable "competence" for specific knowledge sources, moving beyond simple retrieval to deep, adaptive understanding.
A novel policy optimization method bypasses critic action gradients entirely, achieving faster, more robust updates and improved exploration in continuous control.
Human mobility generation can now be achieved without target trajectories by learning relative movement transitions, not absolute destinations, across cities.
LLM concepts form non-linear manifolds, with alignment metrics revealing distinct layer-wise structures and training-dependent multilingual concept sharing that linear methods miss.
A single deep generative model now captures the full joint distribution of mixed-type multivariate outcomes, outperforming specialized methods.
Asynchronous learning communities don't just fragment; they predictably fragment *before* converging around deadlines, a pattern driven by participation shifts, not just behavioral alignment.
Novel dashboards can bridge the gap between complex topological data analysis and user interpretation by explicitly acknowledging and managing methodological uncertainty.
Transformer model explanations diverge significantly under diagnostic uncertainty, highlighting the critical need for multi-method auditing and careful ontology selection in high-stakes medical classification.
Adaptive LLM fine-tuning can achieve superior performance and stability by intelligently searching step-sizes along a trusted gradient direction, costing less than a full line search.
Discrete diffusion models can achieve optimal sampling complexity for categorical MRFs by decomposing the score into time- and target-dependent multiplicative components, allowing for a single, weight-shared score network.
A single denoising process now unifies discrete token generation and continuous latent trajectories, enabling diffusion models to jointly reason about token dependencies and global structure.
A unified proactive generation framework enables LLMs to form reusable factual memories from streaming video without compromising real-time perception.
Retrieval pre-training significantly boosts typed decision model performance for biomedical QA, especially when retaining retrieval priors, but its effectiveness hinges critically on the chosen decision head architecture and training objective.
LLM/VLM agents significantly outperform non-agentic methods in preserving unedited parts during iterative 3D asset modification, but current agentic edits are prohibitively slow.
LLMs can learn to generate syntactically valid text without sacrificing semantic quality, even with underspecified prompts, by leveraging self-supervised rewards derived from their own likelihoods.
Task-specific LoRA-Experts and prototype-ensemble matching unlock rehearsal-free class-incremental learning with reduced interference and improved discrimination.
A single time-series language model now masters forecasting, contextual prediction, and temporal reasoning, outperforming specialized models on key benchmarks.
Language, when self-evolving and treated as a physical representation, can unlock state-of-the-art physical plausibility in video world models, even surpassing proprietary systems.
