Search papers, labs, and topics across Lattice.
100 papers published across 4 labs.
System-aware optimizations in KV cache design could dramatically reduce the memory footprint and costs of serving large language models.
Inconsistencies in Bitcoin and Wikipedia's governance reveal significant flaws in their democratic processes, challenging the very foundations of decentralized decision-making.
MLQENABLER allows secure machine learning queries on encrypted databases, striking a balance between data privacy and ML performance.
PGD-NO breaks the memory barrier for neural PDE solvers, enabling high-fidelity simulations on meshes with over 10 million nodes.
Even with just one-bit communications, agents can perform complex global computations in dynamic networks, challenging assumptions about communication limits.
MLQENABLER allows secure machine learning queries on encrypted databases, striking a balance between data privacy and ML performance.
PGD-NO breaks the memory barrier for neural PDE solvers, enabling high-fidelity simulations on meshes with over 10 million nodes.
Even with just one-bit communications, agents can perform complex global computations in dynamic networks, challenging assumptions about communication limits.
GPU frequency behavior reveals inter-kernel dependencies that could revolutionize latency-prediction models in machine learning.
zkComposer achieves up to 6.84x faster proof generation for zero-knowledge machine learning, revolutionizing the scalability of private model inference.
gspDAG-FL achieves near-optimal learning quality with significantly reduced coordination costs, even in the presence of Byzantine and lazy participants.
Federated learning can significantly boost cardiovascular disease risk prediction accuracy without compromising patient privacy, achieving a C-statistic increase of 0.011 in a challenging data-sharing environment.
Poisoning attacks can severely undermine autonomous vehicle systems, but a new framework effectively filters out malicious influences to ensure safer decision-making.
System-aware optimizations in KV cache design could dramatically reduce the memory footprint and costs of serving large language models.
Balancing session-centric scheduling can boost LLM cluster throughput by up to 16% without sacrificing token reuse.
Predicting user scheduling states can recover up to 73% of the sum rate loss caused by backhaul delays in coordinated beamforming systems.
FedOPAL achieves state-of-the-art accuracy in federated learning without incurring server-side training costs, revolutionizing edge intelligence collaboration.
Even with a majority of compromised keys, adversaries can't forge authorization seals, revolutionizing digital asset custody security.
Naive verification methods for PLCs can produce 44% false alarms due to unrealistic sensor models, but a new hardware-faithful approach eliminates these errors entirely.
DeepCORD adapts solver parameters in real-time, outperforming traditional methods in distributed optimization across complex geometric scenarios.
Achieving near-centralized localization performance in multi-robot systems without the need for extensive communication could revolutionize how teams of robots operate in real-world environments.
Aggressive sampling can significantly enhance performance in asynchronous Kaczmarz methods, but only if tuned just right—too much can lead to instability.
Coded offloading not only boosts performance but also reveals a surprising delay-energy-privacy trade-off that challenges conventional task execution strategies.
Fiber Memory can slash weight-delivery energy by over 70% while eliminating redundant storage across thousands of AI accelerators.
Pre-processing error vectors with Singular Value Decomposition can drastically enhance the efficiency of quantum error correction in distributed systems.
Efficient self-stabilizing algorithms for local symmetry breaking problems can now be executed in a truly uniform distributed model, challenging previous limitations in computational expressivity.
Log-Insight transforms incident diagnosis from a manual, time-consuming process into an efficient automated workflow, achieving over 90% accuracy in identifying root causes within a minute.
Migrating large-model inference to non-GPU accelerators like Huawei Ascend reveals eight critical limitations that could derail performance and reliability.
Mispredictions in power savings can occur, but they never lead to incorrect outputs, making Stochastic Activity Prediction a game-changer for tensor accelerator efficiency.
Quantization errors in speech enhancement models can be largely mitigated by robust spatial filtering, enabling efficient deployment on low-resource devices.
Unifying GPU communication semantics could eliminate the fragmentation in OpenSHMEM implementations across different vendors.
TRM-Raft keeps malicious leader influence under 5% even with a significant presence of Byzantine nodes, offering a robust solution for decentralized systems.
Hidden Decoding achieves unprecedented performance improvements in large language models by scaling computation along the sequence length without modifying the Transformer architecture.
FeLiX slashes the time-to-target accuracy in federated learning by over 2X, making models far more responsive to real-time user data.
Bypassing database drivers can lead to up to 27x speedups in analytical workloads by directly reading storage files with LLM-assisted code synthesis.
Legacy paper ECGs can now be transformed into actionable diagnostic tools in under 30 seconds, even in resource-constrained settings.
Achieving formal privacy in federated learning without sacrificing model performance, FedKT-CSD outperforms traditional methods even under stringent privacy constraints.
Concentrating model capacity on delegation roles can yield substantial performance gains in hierarchical search agents, revealing a critical bottleneck in task decomposition.
Existing vulnerability assessment tools miss critical operating system flaws in software-defined vehicles, exposing a vast landscape of security threats.
Autonomous validator management using fuzzy inference can significantly reduce resource waste and improve blockchain efficiency under varying workloads.
Current Trojan detection methods overlook the stealthy risks posed by compromised standard-cell libraries, revealing a critical vulnerability in zero-trust IC design.
Achieving 99.60% accuracy with a model that requires only 2,370 FLOPS could redefine the landscape of IoT security solutions.
eBIM redefines blockchain infrastructure management by integrating RISC-V to reclaim trust and flexibility lost in traditional BaaS models.
Voters can now securely cast and verify their ballots without relying on a trusted key dealer, dramatically enhancing the integrity of electronic voting systems.
A revolutionary authentication protocol that slashes authentication bandwidth by 99.2% while ensuring complete vehicle-level data sovereignty.
Federated learning can be exploited to encode and extract private training data, revealing a surprising vulnerability in multi-client environments.
Achieving 99.94% of the theoretical maximum for entanglement certification reveals a new frontier in the efficiency of quantum software development.
Voltron boosts LLM accuracy by 16.5% by harnessing the power of multiple edge devices, transforming how we think about local AI execution.
Transforming gradients into a near-isotropic space can cut LLM pretraining time by 7.6% while enhancing downstream task performance.
Achieving up to 31.8% lower latency in multi-GPU systems could redefine how we optimize Large Language Model serving under strict latency constraints.
Achieving efficient multiple double arithmetic on NVIDIA tensor cores could significantly boost performance in high-precision applications.
The cost gap between new AI entrants and established incumbents is set to widen, with incumbents enjoying a 3-4x advantage by 2029-30.
Achieving a 2.52x speedup in quantum circuit execution could redefine efficiency benchmarks in scalable quantum computing.
A miter-aware mapping framework can slash SAT solving times by over 92% by aligning circuit structures with solver needs.
Achieving a 3.5x boost in Energy-Delay-Inverse-Yield, ThermoDSE redefines the optimization landscape for chiplet-based DNN accelerators under thermal constraints.
Achieving over 96% inter-node weak scalability with a grid size of 1 billion cells transforms the capabilities of flow simulations in computational fluid dynamics.
AgentLocate reveals not just which agent failed, but also the critical moment when the system first went off track, outperforming traditional methods in efficiency and accuracy.
Achieving three times the throughput of existing lock-free MCAS algorithms while effectively preventing the ABA problem could redefine concurrent memory access efficiency.
Pre-trained Transformers can effectively detect sophisticated FDI attacks in smart grid communications without the need for complex feature engineering.
A compatibility issue in the EIL SDK reveals significant barriers to stable cross-rollup voucher interoperability, underscoring the fragility of current implementations.
Micro-dependencies can pose outsized risks, with some hiding exploitable vulnerabilities in plain sight due to their extensive network of dependents.
Real-time audio and video traffic can be securely transmitted over a VPN using quantum-derived keys, merging classical and quantum security paradigms.
Thoughtful feature curation reveals that structural code complexity is a far stronger predictor of deployment risk than traditional change volume metrics.
Achieving a 5 to 6 orders of magnitude speedup in effective resistance analysis could revolutionize the design of reliable 3D IC power delivery networks.
SOCI slashes cold-start pull times from 20 seconds to just 2.8 seconds, revolutionizing container deployment efficiency without requiring any changes to existing image formats.
Applications can now enforce their transaction ordering preferences without sacrificing composability, even in adversarial environments.
Estimation, not grid search, is key to optimizing LLM serving—this new approach reveals hidden performance potential in resource management.
Continuous CRS generation in zero-knowledge proofs is now feasible without centralized coordination, even in the face of adversarial challenges.
Achieving unprecedented speedups for solving Poisson's equation in low-rank scenarios could redefine computational limits in scientific simulations.
Direct BRAM-DSP connections can boost FPGA performance by 25% without the need for extensive architectural overhauls.
PRoVeFL achieves unprecedented efficiency in federated learning, improving runtime by up to 100x while ensuring robust privacy and verifiability.
Adversarial backdoor attacks can be mitigated with a defense strategy that ensures attack success probabilities asymptotically approach zero, even with minimal auditing efforts.
Inconsistencies in Bitcoin and Wikipedia's governance reveal significant flaws in their democratic processes, challenging the very foundations of decentralized decision-making.
Game-theoretic reinforcement learning can effectively neutralize attackers manipulating beamforming in 6G integrated sensing and communication systems.
Static metrics fall flat in predicting Java method energy usage, but adding execution time boosts accuracy significantly—up to 0.46 R2.
Achieving a 99.81% logical accuracy in quantum error correction by intelligently routing only 3.3%-6.2% of syndromes to a refinement stage reveals a transformative approach to real-time decoding.
Generating scenarios that directly minimize operational costs can cut grid dispatch expenses by over 2% compared to conventional methods.
A novel error-correction method for QCNNs reduces qubit overhead while significantly improving learning rates in noisy environments.
Energy arbitrage profits in dairy farms can be increased by 18% through a novel multi-agent reinforcement learning approach to battery management.
UBEP slashes All-to-All latency by over 52%, unlocking the full potential of Mixture-of-Experts models on superpod architectures.
Achieving seamless automation in multi-vendor IPoDWDM networks could revolutionize how we manage and optimize complex network infrastructures.
StateFuse reveals that preserving contradictions in memory can enhance safety and correction mechanisms in multi-agent systems, challenging the conventional wisdom of collapsing conflicting information for simplicity.
Manipulating GPU workloads can destabilize power infrastructure, raising current total harmonic distortion to 46.8% and risking cascading failures in data centers.
Trust governance can now evolve seamlessly within decentralized systems, transforming how we manage Byzantine resilience in CRDTs.
Crossroads allows any blockchain asset to seamlessly participate in smart contracts, transforming the landscape of cross-chain finance and asset management.
Achieving GPU performance on par with top CPU-based FHE libraries, LibFHE streamlines the implementation of fully homomorphic encryption without sacrificing efficiency.
Reducing picture buffer access from 152 to 1 could revolutionize the efficiency of video encoding in VVC.
DDB achieves a staggering 100% fault localization success rate, revolutionizing interactive debugging in distributed systems.
MatrixFSDP slashes optimizer-step latency by over 54x while enabling training on models that exceed 80 GB GPU limits.
NPUs can waste up to 40% of energy due to suboptimal configurations, but a new profiling tool reveals how to cut this waste significantly.
TI-dMWF achieves centralized signal estimation in wireless acoustic sensor networks without iterative computation, revolutionizing efficiency in topology-unconstrained environments.
POE2's universal programming model transforms the daunting complexity of advanced memory protection into manageable abstractions, enhancing security without sacrificing usability.
CAP achieves up to 86% higher throughput in MoE models while preserving accuracy, revolutionizing how we optimize expert placement and pruning.
Optimizing multithreading in event generators can drastically cut energy costs, paving the way for sustainable high-energy physics research.
Tail-focused evaluation reveals that traditional methods can misrepresent job runtime predictions, with UserReq outperforming other models in critical scenarios.
Quantum algorithms can't outperform classical methods in 3-coloring rooted trees and 2-coloring even cycles, challenging assumptions about quantum advantages in distributed computing.
ETC slashes migration latency by up to 6.37 times, transforming how LLMs adapt to dynamic resource environments.
RL-driven adaptive batching can yield a staggering 348% throughput improvement in multi-GPU settings, far surpassing traditional heuristics.
Achieving up to 30.9x speedup in ML inference by optimizing workload partitioning between CPUs and CIM accelerators reveals the untapped potential of heterogeneous computing.
Routing anonymity in quantum cloud computing is not just a theoretical concern; it decays exponentially, revealing significant privacy vulnerabilities in noisy quantum hardware.
Achieving up to 4.7x speedup in TGNN training without sacrificing accuracy could redefine performance benchmarks in dynamic graph applications.
Elastic membership in LLM inference can boost throughput by up to 75% compared to static core allocations, all while maintaining bit-exact output.
A novel LRF gateway cuts operational misroutes by over 90% in LLM scheduling, preventing GPU crashes and optimizing resource allocation.
Secure key agreement is possible even in noisy environments, thanks to twin optical PUFs that can withstand fabrication variability.
Tolerating up to 54% of replicas equivocating, this protocol redefines the limits of fault tolerance in state machine replication.
