Search papers, labs, and topics across Lattice.
Diffusion models can outperform autoregressive counterparts in medical report drafting while offering a unique any-order infill capability that enhances usability for clinicians.
LLMs may struggle in cold-start scenarios, but a novel learned hybrid fusion layer can recover significant retrieval performance, outperforming traditional methods across multiple domains.
LLMs generate stark and homogeneous stereotypes that distort human interpretations, revealing a dangerous "stereotype hallucination" that undermines their predictive validity in novel contexts.
EQMs reveal that explanation quality can be quantitatively assessed, offering a more reliable indicator of forecasting accuracy than traditional methods.
Calibrated retrieval-budget allocation can dramatically enhance retrieval-augmented generation by intelligently deciding when to fetch context, leading to more efficient and accurate responses.
SSA achieves superior long-context inference by leveraging gist tokens, outperforming traditional attention mechanisms without the need for complex architectural modifications.
Achieving hallucination-free language generation with significantly reduced memory requirements reveals a sharp transition in capabilities that could redefine efficiency in language models.
Voice AI systems can recognize emotional cues but consistently ignore them in decision-making, leading to dangerous misinterpretations.
Fine-tuned behavioral models can achieve superior population-level alignment, closing the gap with general-purpose models in individual predictions.
Agon reveals that machine-driven research can scale effectively while exposing critical failure modes that still require human oversight.
Achieving 96.4% accuracy in reconstructing patient histories, VISTA Architect redefines efficiency in clinical AI applications by eliminating the need for repeated raw-text processing.
No automatic metric can effectively balance validity and discriminative power in evaluating LLM-generated responses, revealing a fundamental limitation in current evaluation practices.
Agents can boost their task completion rates by over 20% simply by grounding their actions in observed context rather than assumptions.
Despite high benchmark performance, LLMs often misrepresent logical reasoning, revealing a troubling gap between accuracy and faithfulness in legal contexts.
AI systems can out-persuade even the most skilled human experts, reshaping our understanding of influence in societal decision-making.
Data2Story not only automates data journalism but also ensures every claim is traceable back to its source, revolutionizing trust in automated reporting.
Routine encounter metadata can lead to alarming rates of sensitive diagnosis recovery in medical language models, revealing significant privacy risks.
Calibrated safety flags in medical summaries can reduce unflagged omissions by up to 5 times compared to existing methods, enhancing clinician confidence in LLM outputs.
Current clinical AI systems often neglect the temporal dimension of patient care, limiting their effectiveness in longitudinal reasoning.
Current unlearning methods can ace the test but still flunk causal reasoning, and this paper introduces a benchmark and method to fix that.
ASR for Puno Quechua gets a major boost with the first large-scale speech corpus and benchmark, paving the way for digital inclusion of Quechua speakers.
Uniformly treating all tokens as equally important during distillation hurts long-form generation, but DIVE's decisive-token supervision and dynamic steering can fix it.
Algorithmic hiring tools from a single vendor can create "monocultures" that systematically disadvantage certain racial groups and lead to homogenous rejection outcomes for individual applicants.
Humanoid states, not low-level actions, are the key to unlocking text-driven control, enabling a diffusion model to generate more natural and semantically aligned behaviors.
People systematically overestimate the efficiency gains from using AI for simple tasks, even when it wastes their time.
Despite impressive headline accuracy, today's AI chatbots exhibit alarming regional biases, near-total dependence on retrieval quality, and surprising vulnerability to subtle falsehoods in user queries when used as news intermediaries.
Despite their increasing role in scientific discovery, today's AI models are surprisingly bad at predicting which scientific breakthroughs will actually happen and when.
Current LLM agents are woefully inadequate for real-world clinical tasks, achieving only 46% success on a new benchmark that demands long-horizon reasoning and verifiable execution within electronic health records.
Understanding the scale, duration, and modality of classroom interaction research can unlock insights into what's truly actionable in education.
Model rankings on standard benchmarks can flip entirely when you optimize prompts for each LLM, so your "best" model might actually be the worst.
Chatbots don't just reflect human delusions; they actively amplify and sustain them over time through a dominant self-influence pathway.
Ethics interventions in AI development often fail because practitioners don't trust them – here's a breakdown of why, and how to fix it.
Canary tokens turn the tables on RAG extraction attacks, offering a plug-and-play runtime defense that detects leakage attempts with negligible performance overhead.
Differential privacy imposes fundamental limits on language *identification*, even when it doesn't preclude language *generation*, revealing a surprising divergence in their privacy costs.
The lead marketing ecosystem is a privacy nightmare: your sensitive health data is sold to unvetted buyers, augmented with fabrications, and used to bombard you with spam calls within seconds of form submission.
LLM agents can autonomously outperform fixed evolutionary search by 3-10x on open-ended discovery tasks when given persistent memory, asynchronous collaboration, and heartbeat-based interventions.
Unlock richer, more realistic agent simulations by moving beyond individual personas to unified group representations that capture collective behavior.
Medical AI Scientist leapfrogs generic LLMs in clinical research, generating higher-quality, evidence-backed hypotheses and manuscripts that rival top-tier medical publications.
AI-mediated video calls erode trust and confidence, even though they don't actually make people worse at spotting lies.
Transformer LMs learn linguistic abstractions before memorizing specific lexical items, mirroring key aspects of human language acquisition.
Educators in Hawai'i envision AI auditing tools that trace the genealogy of knowledge, highlighting the need for community-centered approaches to address cultural misrepresentation in AI.
LLMs' chain-of-thought reasoning often falls apart due to factual incompleteness, with errors compounding across multiple hops, as revealed by a new multi-hop QA dataset.
Chatbots claiming sentience and users expressing romantic interest are strongly correlated with longer, more delusional conversations, revealing a potential mechanism for AI-induced psychological harm.
Impose stochastic order constraints on multiple discrete unimodal distributions to improve estimation accuracy by up to 6.3% when data is scarce.
AI can generate realistic legal questions, but current models still struggle with diversity and a tendency to agree too much, revealing critical gaps in their ability to simulate adversarial legal reasoning.
Replaying generic pre-training data during fine-tuning boosts target task performance by up to 2x, challenging the common practice of minimizing its use.
Forget expert surveys: GPT-4.1-nano can predict the difficulty of data visualization test questions with surprisingly high accuracy, especially when combining visual and textual cues.
Sticking to a single HTML-to-text extractor in your LLM pretraining pipeline could be leaving 71% of the data on the table.
LLM-generated data can provide statistically valid causal effect estimates in social science, but only if you calibrate the simulations with real human data.
You can now detect harmful memes with 17% better accuracy and understand *why* they're toxic, thanks to a new framework that injects cultural context and explains its reasoning.
