Search papers, labs, and topics across Lattice.
Direct 2D spatial composition with four binding types unlocks unprecedented controllability in poster generation, outperforming prompt-based methods.
Even state-of-the-art LMMs struggle with reasoning-informed visual editing, with the best model only achieving 56.6% accuracy on a new comprehensive benchmark.
Ambiguous object selection in text-guided image editing is solved by directly annotating the canvas, achieving state-of-the-art performance in complex edits.
Most video generation models fail to render text accurately, with even top performers exhibiting a 25% word error rate on a new benchmark designed for this critical oversight.
Append-only memory guarantees attention collapse in long-horizon video reasoning, but actively rewriting a bounded working memory from an external filesystem boosts hard-subset task accuracy by over 20%.
DYAD (DYadic Assistance Dataset), a synchronized multimodal record of human-human assistance during gearbox assembly, is introduced, a linked interaction structure spanning help seeking, intervention choice, execution, and outcome under egocentric and workspace sensing.
Credit-addressable reasoning boosts multimodal geometry accuracy by over 8 points, revealing the critical role of structured learning in complex tasks.
Fast-weight updates in LongVU-TTT enable MLLMs to retain crucial visual context, significantly boosting performance in long video understanding tasks.
Current vision-language models falter in 3D assembly reasoning, struggling with accuracy and complexity in real-world industrial scenarios.
Current video generation models struggle with visual reasoning, with the best achieving only 51% accuracy on a new benchmark designed to probe their capabilities.
Colorization methods can now adaptively handle both modern and historical grayscale images, reducing color artifacts and improving visual fidelity.
MMDiff reveals that multimodal SAEs can be powerful tools for both auditing and steering MLLM behavior, achieving up to 24% reduction in safety attack success rates without compromising performance on visual question answering.
AI-AI co-creation can generate more creative and novel ideas than human pairs, reshaping our understanding of collaborative creativity in AI.
Adaptive execution horizons can boost task success rates by over 11% while reducing runtime overhead in Vision-Language-Action models.
ReToken boosts vision-language model performance by over 20% on key benchmarks while fitting training and inference within a single GPU's memory constraints.
EndoCLIP can match expert endoscopists in classifying lesions, showcasing the transformative potential of linking clinical findings to images in colonoscopy.
Force-based memory enables VLA models to achieve over 80% success in complex manipulation tasks, outperforming traditional memory methods with minimal computational cost.
Transitioning from 2D to 3D modeling reveals that fine details in monocular geometry can be captured with unprecedented fidelity.
ReViV reconstructs 4D viewer and view dynamics from a single monocular video, achieving unprecedented accuracy and speed without heavy task-specific priors.
SciForma achieves unprecedented structural fidelity in scientific diagrams, outperforming both open-source and proprietary models.
Natural paper revisions can be harnessed to train AI agents for precise and context-aware editing of complex scientific diagrams.
Open-AoE transforms egocentric video capture into a powerful resource for embodied intelligence, making it easier than ever to train robots with human-like manipulation skills.
Current MLLMs miss critical insights from multi-view sports videos, but a new agentic framework boosts performance by over 14% through smarter view selection.
MedPMC transforms millions of clinical articles into a goldmine of high-fidelity image-text pairs, dramatically improving model performance and clinical relevance.
VLMs are prone to critical failures that vary significantly across cultures, exposing the inadequacy of Western-centric safety benchmarks.
CAIRN redefines 3D scene understanding by seamlessly integrating room-level topology with object-level relations, achieving unprecedented performance in multi-room environments.
UNIVERSE achieves a remarkable 4.3× speedup in trajectory inference while maintaining planning accuracy, revolutionizing how video dynamics inform autonomous driving actions.
ResearchStudio-Reel not only automates research dissemination but does so with unprecedented quality, outperforming both traditional methods and leading LLMs in aesthetic appeal and information accuracy.
Interleaving speech and text during ASR training boosts entity recognition accuracy and narrows the gap between modalities, challenging traditional training paradigms.
Achieving 100% task success in closed-loop execution, Embodied.cpp revolutionizes how embodied AI models are deployed across diverse hardware platforms.
Ink3D achieves a breakthrough in 3D asset creation, enabling the generation of complex textures that were previously unattainable with conventional methods.
PRIME-Speech achieves low-latency, accurate speech-to-speech generation without sacrificing the robust performance of existing speech-to-text models.
Bridging the Context Gap in T2I models, Qwen-Image-Agent achieves state-of-the-art performance by intelligently constructing context from user input and external sources.
Robots can now navigate crowded spaces more effectively by understanding human intentions, thanks to a new method that integrates rich visual cues into their decision-making process.
MambaRaw achieves a remarkable 1.4 dB increase in PSNR at low metadata bitrates while slashing coding latency by nearly 9%, setting a new benchmark in raw image reconstruction.
High-diversity training improves safety in VLA models, but sub-optimal trajectory synthesis still hinders task success.
Zero-shot transfer of a refined RL policy boosts manipulation success rates from 42% to 76% on real robots, showcasing a breakthrough in sim-to-real applications.
MuseVLA achieves an impressive 80.6% success rate in robotic manipulation tasks by leveraging diverse sensing modalities, surpassing traditional RGB-only models.
CoTriSyGen achieves unprecedented long-range coherence in video generation by integrating visual evidence into a dynamic memory system, drastically reducing identity drift across shots.
Retrieving the right prompts can boost LMM performance by up to 30%, challenging the assumption that similarity guarantees effectiveness in in-context learning.
A million-scale dataset for identity-preserving video generation enables a new benchmark that outperforms existing models with minimal parameter overhead.
Code-based 3D reconstruction achieves superior edit fidelity and locality, outperforming traditional point-cloud methods in preserving unedited regions.
Latent spatial memory can accelerate video generation by over 10 times while dramatically reducing memory usage, revolutionizing how we model dynamic scenes.
CapRL++ redefines caption quality through utility, enabling models to produce high-fidelity descriptions without the constraints of traditional supervised fine-tuning.
Encoder-free speech modeling can rival traditional methods, challenging the necessity of dedicated speech encoders in LLM architectures.
AsyncWebRL achieves a staggering 2.9× increase in training throughput while setting a new state-of-the-art performance for web agents on challenging tasks.
OpenWebRL-4B sets a new benchmark for open-source visual web agents, achieving impressive success rates with minimal initial data while outperforming larger-scale competitors.
Optimizing for runtime in multimodal training can be energy-inefficient, as data movement and overlap on Grace Hopper chips dominate energy consumption, not raw compute.
Forget grid layouts: Map2World lets you generate consistent 3D worlds from arbitrary segment maps, offering unprecedented control and scalability.
