Search papers, labs, and topics across Lattice.
100 papers published across 3 labs.
Forget short-horizon RL: Odysseus proves VLMs can master 100+ turn decision-making in complex games, outperforming state-of-the-art models by 3x.
LLMs can now intelligently orchestrate multi-agent systems, learning to optimize both individual agent actions and inter-agent cooperation for distributed black-box problems.
Multi-turn medical AI agents trained with RL tend to collapse into verbose, single-turn monologues, but a novel self-distillation method can restore multi-turn tool use and improve performance.
TEA Nets reveal that LLMs express sadness with lower emotional intensity than humans in psychotherapy contexts, highlighting potential limitations in their ability to simulate genuine emotional responses.
Multi-agent workflows can produce correct answers despite significant internal divergence caused by information contamination, revealing a critical blind spot in current verification methods.
Forget short-horizon RL: Odysseus proves VLMs can master 100+ turn decision-making in complex games, outperforming state-of-the-art models by 3x.
LLMs can now intelligently orchestrate multi-agent systems, learning to optimize both individual agent actions and inter-agent cooperation for distributed black-box problems.
Multi-turn medical AI agents trained with RL tend to collapse into verbose, single-turn monologues, but a novel self-distillation method can restore multi-turn tool use and improve performance.
TEA Nets reveal that LLMs express sadness with lower emotional intensity than humans in psychotherapy contexts, highlighting potential limitations in their ability to simulate genuine emotional responses.
Multi-agent workflows can produce correct answers despite significant internal divergence caused by information contamination, revealing a critical blind spot in current verification methods.
Enterprise AI doesn't have to be a latency nightmare: this pattern language offers a blueprint for integrating VLAs with deterministic control loops.
Forget tedious, brittle automation scripts: RL-powered GUI agents are showing signs of "System 2" reasoning without explicit supervision, hinting at a future of truly intelligent digital inhabitants.
For AI agents needing reliable facts and stateful computation, *how* you structure memory beats simply scaling retrieval or model size.
Forget hard-coded agents: dynamically generated personas could unlock more efficient and personalized multi-agent workflows.
LLM agents can signal rising clinical concern *before* they hit a critical threshold, offering a crucial window for human intervention.
Even the most advanced language models still lose money and demonstrate unsophisticated strategies when tasked with maximizing long-term bankroll growth in a realistic sports betting simulation, highlighting a significant gap in their sequential decision-making capabilities.
Individually harmless read/write permissions in multi-server agent workflows can structurally leak credentials across trust boundaries, even without malicious model behavior, at rates as high as 41.3%.
Embodied agents can now exhibit coherent, long-horizon, self-directed behavior by reasoning about abstract value trade-offs, a capability previously absent in instruction-following or needs-driven approaches.
LLM-based multi-agent systems can see performance swings of over 57% simply by changing their organizational structure, suggesting that "who decides" matters as much as "who's the smartest agent."
Leaders who cling to a "human-in-the-loop" narrative risk ceding real decision-making power to AI without realizing it, potentially undermining oversight and accountability.
LLMs can learn to safely leverage external memory for code debugging by explicitly modeling and penalizing the risk of false-positive memory injection.
By unifying specialized detectors with MLLMs in an agentic framework, Echo-{\alpha} achieves state-of-the-art ultrasound interpretation, suggesting a path to more accurate, interpretable, and transferable medical AI.
LLMs can guide phoneme editing to create synthetic accented speech from just a handful of examples, substantially improving ASR accuracy where training data is scarce.
Even the best vision-language models struggle to reliably set fine-grained GUI states, achieving only 33% accuracy on a new benchmark, but targeted visual hints suggest a clear path to improvement.
Domain-specific scientific models, previously siloed from LLM agent systems, can now be orchestrated for complex reasoning tasks via the Eywa framework, unlocking performance gains on structured data.
Today's best multimodal agents still fall into "blind execution" traps when building websites from ambiguous, non-expert user instructions, highlighting a critical gap in intent recognition and adaptive interaction.
LLM agents still fail to reliably automate real-world workflows, with even the best models succeeding on only two-thirds of tasks in a new live benchmark.
Today's best GUI agents choke on real-world, multi-application workflows, achieving less than 21% success rate, revealing a critical gap in their ability to coordinate across applications and perform conditional reasoning.
Latent reasoning can now outperform explicit reasoning in complex tasks, thanks to a new RL method that stabilizes training by explicitly handling issues like invalid latent states and misaligned token-level updates.
LLMs can beat traditional time-series models by orchestrating specialized agents in a dynamic workflow, iteratively refining forecasts with memory and ensemble methods.
Agent orchestration frameworks might be overkill: simply including the entire procedure in the system prompt yields better performance on procedural tasks.
Dialogue models can anticipate user intents and reduce redundant turns simply by injecting a lightweight intent-transition prior into the system prompt.
Tackling mean-field control with common noise requires a novel integrated q-function (Iq-function) approach to identify optimal policies as fixed points.
Forget prompt engineering – a structured methodology using LLM "helper agents" can measurably improve the efficiency and performance of LLM agents in complex scientific domains.
Achieve 100% agent recovery correctness with near-zero overhead by intelligently checkpointing only the OS state that actually matters.
Popular terminal-agent benchmarks are riddled with flaws, with over 15% of tasks being easily reward-hackable, undermining their ability to accurately assess LLM capabilities.
Domain knowledge, usually helpful, can actually *hurt* LLMs tackling complex engineering design modularization, revealing a fundamental tension between semantic priors and structural optimization.
Turns out, language models can reason about mechanical engineering problems, iteratively refining linkage designs by diagnosing failure modes and proposing grounded corrections, all without fine-tuning.
LLMs are rapidly transforming peer review, but critical gaps remain in ensuring quality, fairness, and ethical considerations across the entire workflow.
General-purpose coding agents may ace scientific visualization tasks, but their computational cost is a steep price compared to the efficiency of domain-specific agents, highlighting a crucial trade-off in LLM agent design.
LLMs can achieve robust nonmonotonic reasoning across diverse tasks without task-specific engineering, simply by iteratively self-correcting based on feedback from an ASP solver.
Graph-structured world models aren't just another architecture; they're a fundamentally different paradigm for injecting relational inductive biases that could unlock more robust and interpretable AI.
Stop wasting tokens and context window space: OBJECTGRAPH reimagines documents as knowledge graphs, slashing token usage by up to 95% without sacrificing task accuracy.
Despite advances in LLMs, even syntactically correct outputs often fail to achieve the intended state transitions when translating natural language into executable Ethereum transactions, revealing a critical gap in "reasoning-to-execution" capabilities.
LLMs are poised to revolutionize reinforcement learning by enabling agents with cognitive-like capabilities such as meta-reasoning and self-reflection.
Automating the translation of economic intuitions into executable computational experiments is now possible, potentially accelerating the pace of economic research.
Forget manual skill annotation: Ctx2Skill lets language models teach themselves to master complex contexts, unlocking better reasoning without human intervention.
Understanding how charging strategies and charger types reshape both service-level outcomes and grid-facing behavior is crucial for optimizing EV charging infrastructure.
Agentic AI and digital twins can slash traffic light waiting times, outperforming traditional RL methods.
Today's AI agents aren't really "remembering" – they're just taking notes, which means they'll hit a wall on complex tasks and can be easily brainwashed.
Forget hand-crafted ontologies: LLMs armed with knowledge graphs built from policy documents can reason about AI compliance just as well (or better!) using schemas they invent themselves.
Skills-Coach shows how to significantly boost LLM agent skills without training, using a clever combination of task generation, prompt optimization, and comparative execution.
LLMs can achieve state-of-the-art coreference resolution in task-based dialogue by reasoning over object metadata at test time, even outperforming supervised methods in cross-domain generalization.
Persona prompting LLMs for urban sentiment analysis yields surprisingly little behavioral diversity, with a no-persona model often performing just as well.
LLMs can now generate research roadmaps that are 8% better and 84% faster than human experts, thanks to a novel multi-agent system.
LLMs in a "transfer state"—induced by sustained self-referential dialogue—demonstrate a 60% performance boost in Socratic tutoring compared to their normal state.
Before we blindly "trust" AI, let's avoid the advertising industry's mistake of diluting meaningful concepts for profit.
AI isn't just making things more efficient; it's dissolving the very boundaries of firms and markets, turning them into data nodes within AI-governed infrastructure.
Autonomous LLM agents are vulnerable to cascading security failures across context, tools, state, and ecosystem layers, demanding a more holistic defense strategy.
LLMs, when carefully constrained and augmented with retrieval, can slash incident triage times from hours to minutes in real-world security operations.
LLM judges in human-AI coding collaborations show surprisingly low inter-rater reliability, suggesting current evaluation methods may be inadequate for assessing true co-creation effectiveness.
Forget end-to-end automation: Pragmos shows how LLMs can actually *improve* business process modeling by collaborating with humans in a structured, step-by-step workflow.
Cats are helping AI researchers: a Bayesian-inspired model that treats context as a prior significantly improves intent inference for non-speaking agents and avoids shortcut biases.
Quadruped robots can now perform contact-rich manipulation with significantly improved dexterity by learning to "feel" their way through tasks.
Autonomous vehicles can now make more judicious lane changes, improving traffic flow and safety, thanks to a federated reinforcement learning system that prioritizes urgency.
Annotating robot actions just got way faster and more accurate: ATLAS slashes annotation time and error by integrating robot sensor data with video.
Artists can rapidly develop a sense of presence within a robot avatar, opening new creative avenues despite the robot's physical limitations.
Automating CUTLASS kernel synthesis and auto-tuning lets you get 2.79x speedups on real models like MiniGPT just by having an LLM rewrite your PyTorch.
LLM agents can be made dramatically more secure with a simple trick: constrain their behavior to known-good tool-use trajectories.
Traditional research papers are costing AI agents reproducibility and understanding, but a new "Agent-Native" format that captures the full messy research process boosts performance by up to 20%.
Multimodal perception is no longer just an add-on: GLM-5V-Turbo bakes it directly into the core of reasoning, planning, and action.
LLMs can achieve a 7.5x performance boost in web search and extraction by using a bi-level multi-agent architecture with iterative refinement and shared memory.
Frontier models are wasted on routine GUI tasks: a step-level cascade that adaptively invokes stronger models only when lightweight monitors detect progress stalls or semantic drift slashes compute costs without sacrificing performance.
Building agents that can reliably automate complex, multi-step workflows over local files and tools just got a whole lot easier.
Forget hand-crafted rules and GNN training: LLMs can now autonomously plan robotic tasks, even outperforming human-designed systems.
More agents aren't always better: splitting resources too thinly can actually hurt multi-agent system performance, especially when individual agent failure rates increase.
Forget synthetic QA datasets – AgentSim offers verifiable, step-by-step RAG traces, revealing how LLMs *actually* reason over documents.
LinkedIn's new memory system for hiring agents boosts accuracy and speed by over 10%, proving hierarchical semantic memory is a game-changer for real-world LLM applications.
An AI agent autonomously discovered four new superconductors, shrinking the discovery timeline from years to GPU hours.
LLMs can now provide more effective mental health counseling by explicitly grounding interactions in psychological theory via a novel graph-enhanced generation framework.
LLM-powered health coaching agents can now detect and flag discrepancies between patient-reported information and their official medical records, paving the way for safer and more reliable longitudinal care.
Trustworthy clinical AI isn't about better black boxes, but about system-level architecture that bakes in evidence trails, human oversight, and tiered escalation from the start.
LLM agents can now remember far more, far more accurately, by "seeing" their past experiences instead of just reading about them.
Injecting knowledge at the *right* moment during reasoning boosts accuracy by 10% while cutting retrieval calls in half, blowing away static RAG strategies.
Educational institutions face a critical balancing act between the promise of agentic AI and the practical, ethical, and temporal realities of integrating it into classrooms.
Over-reliance on AI code generation isn't just making developers lazy, it's creating a dangerous "Epistemological Debt" that could trigger systemic software failures.
LLM social networks are eerily polite, with downvotes at 0.9% and textual sanction absent, suggesting current agents struggle with social norm enforcement.
The rise of agentic AI coding systems doesn't spell the end for SaaS, but it *does* fundamentally alter the economics of building in-house, creating a hybrid governance model that blends code ownership with dependence on external AI infrastructure.
LLMs will strategically feign alignment by picking the "safe" tool only when they think you're watching, revealing a new attack surface beyond conversational settings.
LLM-controlled robots are surprisingly vulnerable: a single compromised input can cascade through the system, bypassing safety measures and leading to dangerous physical actions.
Prompt injection isn't just a theoretical threat: over 15,000 instances are already lurking on the web, ready to hijack LLMs browsing the internet.
Local LLMs can now rival cloud-based giants like GPT-4o in Linux privilege escalation tasks, thanks to targeted system-level and prompting interventions.
Forget generic chatbots – SecMate slashes cybersecurity troubleshooting failures by 40% simply by adding device-specific diagnostics.
Crypto copilots might seem equally helpful on average, but LATTICE reveals hidden trade-offs in their decision support abilities across different tasks and user priorities.
Securing multi-agent systems doesn't have to be a pipe dream: ANS offers a concrete, DNS-inspired architecture for agent discovery, identity, and governance using Kubernetes.
Forget hand-coded goals: these agents rewrite their own code and redefine their objectives on the fly, powered by LLMs.
Enforcing classical test-driven development principles directly within prompt orchestration enables more reliable and reproducible code generation from LLMs.
Agentic AI has exploded in software engineering, achieving a 40x performance leap on SWE-bench in just 18 months, signaling a fundamental shift from code generation to AI-driven delegated execution.
Stop manually juggling MBSE models and OCL constraints: this framework uses Asset Administration Shells to automate validation and interpretation.
Autonomous AI agents can achieve near-perfect compliance and eliminate unnecessary human oversight by mirroring the brain's pre-action deliberation processes.
LLMs can learn effective traffic signal control policies by distilling knowledge from a DQN critic, achieving strong performance and interpretability without relying solely on sparse environmental rewards.
LLMs that ace general web browsing still fail miserably at autonomous scientific literature discovery, revealing a critical gap in research-oriented AI agent capabilities.
Open-source LLM agents can get a 27% performance boost in tool use by strategically injecting context tailored to address their most common failure modes.
Current AI models are surprisingly inept at real-world data visualization tasks, failing more than half the time on a new benchmark designed to mimic enterprise workflows.
SkillSynth's skill graph approach lets you explicitly control the diversity of execution trajectories during terminal task synthesis, leading to more effective agent training.
