Search papers, labs, and topics across Lattice.
100 papers published across 6 labs.
Neural Cellular Automata, blending Wolfram's recursive programs with neural networks, offer a fresh perspective on modeling complex, self-organizing systems.
Finally, a plugin framework that lets you mix-and-match KV-Cache, LoRA, and other controls to steer diffusion models without being locked into a specific backbone.
Multimodal models can now achieve state-of-the-art performance in real-world tasks like document understanding and audio-video comprehension with significantly reduced inference latency thanks to novel token-reduction techniques.
State-of-the-art shot boundary detection gets a major upgrade with a Transformer-based approach that not only improves accuracy but also offers more interpretable boundaries, thanks to a novel relational prediction framework and synthetic training data.
Ditching the vision encoder actually *improves* multimodal understanding at scale, proving that pixel embeddings alone can achieve state-of-the-art results in unified multimodal models.
Neural Cellular Automata, blending Wolfram's recursive programs with neural networks, offer a fresh perspective on modeling complex, self-organizing systems.
Finally, a plugin framework that lets you mix-and-match KV-Cache, LoRA, and other controls to steer diffusion models without being locked into a specific backbone.
Multimodal models can now achieve state-of-the-art performance in real-world tasks like document understanding and audio-video comprehension with significantly reduced inference latency thanks to novel token-reduction techniques.
State-of-the-art shot boundary detection gets a major upgrade with a Transformer-based approach that not only improves accuracy but also offers more interpretable boundaries, thanks to a novel relational prediction framework and synthetic training data.
Ditching the vision encoder actually *improves* multimodal understanding at scale, proving that pixel embeddings alone can achieve state-of-the-art results in unified multimodal models.
PINNs offer a promising new approach to solving complex problems in differential geometry by directly minimizing differential functionals.
The secret to effectively pruning LLMs might not be *how* you search for redundant layers, but *what* you're optimizing for.
Achieve dynamic regret bounds for online regression in RKHS by combining discounted VAW with finite-dimensional subspace approximations, offering a practical approach for time-varying comparisons.
Learning generative models for high-dimensional matrices doesn't have to be a computational nightmare: CoreFlow achieves state-of-the-art results in low-data regimes by learning shared low-rank structure.
Concept extraction's identifiability problem just got a lot easier, thanks to a new framework that turns guarantee proofs into set intersection problems.
Forget training from scratch: HyLo lets you breathe new (long-context) life into your existing Transformer LLMs, achieving 32x context extension and 90% KV-cache reduction.
Multi-anchor word embeddings, previously impractical for LLMs, can now outperform standard embeddings with 98% fewer parameters and a 40x smaller embedding layer.
Not all layers are created equal: pruning the KV cache in a layer-dependent manner significantly boosts long-context LLM performance compared to uniform pruning strategies.
LLMs re-rank documents better when you learn to route each query to the specific attention heads that matter, instead of relying on static subsets or everything at once.
Ditch the complexity of Inter-Blockchain Communication: this tree-based blockchain framework lets you navigate hard forks like directories in a file system.
Quantum-safe certificates bloat TLS handshakes so much that they measurably degrade web performance, and current CDN optimizations aren't enough to fully compensate.
Forget activation counts – RVC slashes Rowhammer mitigation overhead by up to 99.99% by directly tracking a row's vulnerability to bit flips.
Forget complex side-channel analysis: a single, machine-checked theorem proves that masked Barrett reduction leaks at most *one bit* of information per wire, offering a universal security guarantee for post-quantum crypto.
Guaranteeing atomicity in secure enclaves doesn't have to break real-time OS timekeeping – a secure-driven synchronization mechanism can unobtrusively keep everything in sync.
LLMs can now provide interpretable anomaly diagnoses in industrial control systems by translating detection evidence into actionable hypotheses for operators.
Transformer-based vulnerability detection is booming, but this review reveals critical gaps in data balance, interpretability, and cross-language generalization that could be holding back truly robust systems.
Automating monolith-to-serverless migration is now possible with an LLM-powered pipeline that outperforms commercial tools.
LLMs can now generate reliable hardware reference models with 95% accuracy thanks to a novel co-evolutionary verification mechanism that weeds out correlated hallucinations between model and testbench.
Autoregressive image models can now compete with diffusion models in image quality and efficiency, thanks to a variable-length tokenization scheme that decouples compute from resolution.
Achieve real-time, accurate image reconstruction from sparse Laplacian fields using a wavelet neural network with only 200 parameters.
Imagine buildings that adapt to the materials available, not the other way around: this framework uses robots to make it a reality.
Forget clunky animation pipelines – MotionBricks lets you assemble real-time, high-quality character motions like LEGOs, even controlling robots.
Ditch silicon bottlenecks: a novel optoelectronic correlator uses cold atoms to accelerate 3D CNNs by orders of magnitude.
FPGA CAD tools waste enormous time re-checking the same cluster packings, but a simple memoization trick can slash runtime by up to 29x.
Forget A100s for long-context LLMs – Salca achieves up to 74x better energy efficiency with a sparsity-aware hardware accelerator.
Vib2Conf achieves unprecedented accuracy in identifying 3D molecular conformations from vibrational spectra, even distinguishing between near-isomeric conformers differing by only ~1 Å RMSD.
Unlock spectroscopic and electronic observables in large-scale molecular simulations by learning the electron density directly, paving the way for more comprehensive and transferable machine-learned interatomic potentials.
Sub-linear attention is now possible without sacrificing complete long-range dependency retention, thanks to learnable summary tokens that compress context.
Generative recommendation gets a boost: modeling behavior intensity and transitions yields 15-23% gains in prediction accuracy.
LLMs can denoise sequential recommendations by disagreeing with the recommendation model itself, leading to more robust performance against noisy user data.
Achieve millisecond-level 3D point cloud reconstruction from a single image without sacrificing quality, blowing past diffusion model latency.
Concept bottleneck models can now distinguish between reducible model uncertainty and irreducible input ambiguity, enabling targeted interventions like data collection and human review.
Forget rigid multi-agent pipelines: this framework lets you build self-organizing AI "companies" that dynamically recruit talent and adapt to tasks on the fly.
Finally, a TTS system that lets you control the *exact* timing and pauses of individual words, opening the door to applications like perfectly paced guided reading and accessible code narration.
SAD offers a surprisingly fast and accurate alternative to neural implicit representations for image compression and differentiable rendering, achieving 4-19x training speedups while outperforming state-of-the-art methods like Image-GS.
Training a single model across text, images, video, 3D geometry, and hidden representations unlocks "Context Unrolling," where the model reasons across modalities to improve reasoning fidelity.
Explicitly conditioning neural surrogates on supersaturation dramatically improves their accuracy in simulating crystal growth dynamics compared to implicit inference, especially with limited data.
Signal processing offers a surprisingly effective lens for understanding and improving LoRA, the reigning champ of parameter-efficient fine-tuning.
Quotient-space diffusion elegantly sidesteps the need to learn symmetry transformations, leading to more efficient and accurate generative models for systems with inherent symmetries.
Solve new PDEs 100x faster with 10x less error by learning a transferable PINN representation and adapting to new equations with a single closed-form calculation.
LLMs can now reason across long conversations without breaking the bank: StructMem slashes token usage and API calls while boosting temporal reasoning.
Forget philosophical debates: a practical "learning mechanics" is crystallizing to explain *how* deep learning works, not just *why* it should.
Despite their architectural differences, Transformer-based genome language models can provide equally reliable biological insights as CNNs, as revealed by attention-based explainability methods.
N-gram models can rival neural networks in event log prediction, but the secret sauce is a smart ensemble method that dynamically promotes the best model during inference.
Forget ReLU's rough edges: a new family of smooth activation functions, GEM, closes the gap with GELU and even outperforms it in some cases, revealing a surprising architecture-dependent sweet spot for smoothness.
Achieve state-of-the-art periodic signal denoising with a single, lightweight dilated CNN that generalizes across frequencies via resampling.
IoT intrusion detection gets a boost: A-THENA's time-aware encoding and network-specific augmentation beats state-of-the-art methods by up to 6.88% in accuracy, all while running on a Raspberry Pi Zero 2 W.
Volatile memristors can achieve state-of-the-art image classification accuracy in reservoir computing, even with significant device variability, suggesting they are a viable alternative to traditional CMOS.
LSTMs can bring low-cost air quality sensors up to regulatory compliance, unlocking dense urban monitoring networks previously limited by calibration challenges.
ResGIN-Att's cross-attention mechanism not only boosts drug synergy prediction but also offers a peek into the "why" behind drug interactions by highlighting crucial chemical substructures.
Achieve LLM personalization with the guarantee that deleting a small user-specific proxy deterministically erases all traces of their data, sidestepping the need for computationally expensive retraining.
VARestorer distills a text-to-image VAR model into a one-step super-resolution network, achieving state-of-the-art image quality with a 10x speedup.
PINNs can now efficiently solve highly oscillatory wave equations in heterogeneous media, thanks to a Green's function-based integral formulation that cuts computation by 10x and avoids absorbing boundary layers.
Forget compressing entire tokens – selectively routing *parts* of tokens based on query relevance unlocks better compression-quality tradeoffs in LoRA-adapted transformers.
Cross-entropy loss isn't just a detail – it's the unsung hero behind how well energy probes work in predictive coding networks, accounting for up to 66% of the probe-softmax gap.
Channel-free HAR is now possible: a single model can perform activity recognition across diverse IoT sensor setups without needing fixed channel arrangements, thanks to metadata-conditioned fusion.
A game-theory-inspired ensemble of LLMs and a lightweight verifier slashes the cost of code vulnerability detection while boosting accuracy, proving that strategic agent design can beat brute-force scaling.
Halving the parameter count of LLMs without sacrificing performance is now possible with Hyperloop Transformers, thanks to looped layers and hyper-connected residual streams.
Vector-based fine-tuning just got an 8x speed boost, rivaling LoRA's performance with a fraction of the parameters, thanks to a clever gradient-informed initialization.
Autoregressive video diffusion models can achieve faster decoding, lower memory footprint, and higher quality long-horizon generations by learning to attend to only the most salient spatiotemporal blocks.
Forget repeatedly re-running inference on residual graphs: this GNN-guided Ford-Fulkerson algorithm learns edge importance probabilities to dramatically accelerate max-flow computation and image segmentation.
Recurrent Transformers let you trade model depth for width, slashing KV cache memory footprint and inference latency without sacrificing performance.
A transformer-based deep learning approach can not only drastically accelerate Unit Commitment problem-solving but also, surprisingly, find lower-cost operational schedules than traditional MILP solvers in certain instances.
LLM agents are wasting up to 60k tokens per turn on unnecessary tool schemas – Tool Attention slashes this "Tools Tax" by 95% and unlocks truly scalable agentic workflows.
Achieve competitive video copy detection accuracy with descriptors orders of magnitude smaller and inference speeds exceeding 11k samples per second by replacing floating-point operations with a learned Boolean circuit.
Unseen token generalization in transformers isn't just about copying; it's fundamentally limited by a representational collapse in the unembedding space.
Securing energy grids against cyberattacks may hinge on clever observer/controller architectures that respect data privacy and regulatory constraints.
Autonomous vehicles can now plan trajectories 10x faster without sacrificing performance, thanks to a novel architecture that learns complex driving behaviors in latent space during training.
Hybrid architectures that combine attention and recurrence can maintain reasoning performance as task complexity increases, while transformers see a sharp performance drop-off.
MemPalace's impressive memory recall isn't due to its fancy "memory palace" spatial organization, but rather its simple "store everything verbatim" approach combined with a strong embedding model.
Automate resource management in Linear Haskell with linear constraints, eliminating the need for explicit linear arguments and streamlining development.
Forget generating static shapes – Sculpt4D now lets you efficiently sculpt dynamic 4D objects with state-of-the-art temporal coherence.
Guarantee application-level protocol compliance without touching application code by pushing runtime verification into the network itself.
Achieve a 10x speedup in detecting tiny objects in massive satellite images without sacrificing accuracy, even on a single GPU.
Face forgery detectors crumble when evaluated on unseen data, but a new metric, Cross-AUC, finally exposes this hidden vulnerability.
Achieve more precise facial attribute editing by decoupling attribute manipulation from image synthesis, sidestepping the optimization challenges of directly combining GANs and diffusion models.
Steal accuracy from dense models and stabilize MoE training with a simple teacher-guided routing scheme that combats gradient starvation.
Achieve near-perfect brain tumor classification with a Vision Transformer, unlocking clinically interpretable insights via attention rollouts.
Vision GNNs can achieve near 100x speedups on FPGAs by decoupling graph construction from feature updates, enabling concurrent execution without significant accuracy loss after fine-tuning.
By explicitly addressing often-overlooked fusion erasure errors, this new compilation scheme unlocks exponentially more robust photonic quantum computations.
Unlock real-time, high-quality 3D scene reconstruction from unconstrained images with varying lighting, thanks to a feed-forward Gaussian Splatting model that learns appearance embeddings.
Explicitly constraining action generation with predicted spatial "corridors" boosts VLA model performance by up to 12.4% on challenging robotic manipulation tasks.
Bridging the gap between blockchain research and real-world deployment requires navigating recurring design tensions like scalability vs. security, decentralization vs. governance, and privacy vs. compliance.
On-device LLM inference gets a massive speed and energy boost by adaptively streaming only the most expensive parts of the KV cache from the cloud.
Forget hand-tuning: SPAC automatically generates FPGA-based network switches that slash latency by up to 38% while dramatically reducing resource usage.
Achieve state-of-the-art sequential recommendations by aligning multi-resolution temporal dynamics with graph propagation at matching scales.
Tired of opaque elliptic curve parameters? ECCFROG522PP offers a fully transparent and reproducible 522-bit alternative, letting you independently verify its security.
Uncover hidden attack patterns in automotive networks by combining formal verification with process mining, revealing root causes of security vulnerabilities that traditional methods miss.
Forget specialized prefix-parsing algorithms: a simple grammar transformation lets you use standard parsers for efficient prefix parsing and next-token prediction.
Forget text-only pre-training: training on music *first* can dramatically accelerate language learning in small language models.
By dynamically injecting frequency-aware n-gram features, X-GRAM achieves state-of-the-art accuracy with smaller embedding tables, offering a practical path to scaling memory-augmented architectures.
By spectrally decoupling robot control into intent and dynamics, ResVLA offers a more efficient and robust approach to generative VLA policies.
LLMs can rewrite bad job descriptions and category-aware MoEs can better match candidates, leading to a 19.4% boost in recruitment click-through rates and millions saved.
Skip the pixel-perfect annotations: attention-based MIL with pathology foundation models can predict lung cancer growth patterns from whole slide images with surprisingly high accuracy.
By unifying generative and discriminative approaches, UniGenDet achieves superior image generation and detection, suggesting that these tasks benefit from a symbiotic relationship previously hindered by architectural divergence.
