Directory
Papers
Reviewed resources in Papers. 148 resources listed.
LensVLM
arxiv.orgA 9-billion-parameter vision-language model, released with its evaluation code, that scans compressed page images and selectively expands only the relevant ones to their uncompressed form using learned tools, for answering questions over long documents.
Quant
github.comResearch code implementing least-squares binary quantization of neural networks in PyTorch, with configs and scripts for reproducing the paper's MNIST, CIFAR-100 and ImageNet results.
Awesome LLM Agent Papers
github.comAn annotated, actively updated reading list of over 500 papers on LLM agents, organized by function such as planning, memory, tool use and safety, built as a companion to a survey paper.
Research code for teaching a multimodal language model to recognize and remember user-specific concepts through retrieval augmentation, so it can discuss a person's own subjects in later conversations.
Speaking While Listening
arxiv.orgA survey and curated list of full-duplex spoken dialogue systems, the models and architectures that let a voice assistant listen and speak at the same time instead of taking strict conversational turns.
A survey and curated paper list on audio-visual intelligence, covering how foundation models jointly perceive, generate and interact using both sound and vision.
A research method for adapting true 1-bit large language models after training, using sparse XOR bitmask patches that flip individual binary weights instead of requiring gradients or continuous weights.
Research code for a two-stage reinforcement fine-tuning method that trains large language models to reason step by step about time series data before producing a forecast.
PSearch
arxiv.orgResearch code for a search-based patch generation method for large-language-model automated program repair, using Monte Carlo tree search and beam search to explore candidate fixes.
Research code for a reinforcement-learning method that trains reasoning models with dense token-level supervision by contrasting teacher-student gaps under correct and incorrect reference solutions.
The official research codebase for a training method that stabilizes distilling a large teacher agent's behavior into a smaller student agent across long, multi-turn tasks.
Awesome VLA & WAM Quantization
github.comA curated, categorized list of over 180 research papers on quantizing Vision-Language-Action and World-Action models so embodied robot policies fit within memory, latency and energy limits.
A research system, presented at OSDI 2025, for running large language model inference on wafer-scale AI accelerators instead of GPU clusters.
AdaFlash
arxiv.orgThe official implementation of a speculative-decoding method that pairs a large language model with an adaptively trained diffusion drafter model to speed up inference.
MAS-Zero: Designing Multi-Agent Systems with Zero Supervision
vincent950129.github.ioA research framework in which a meta-agent designs, evaluates, and verifies multi-agent system architectures for a task without labeled supervision or a validation set.
LM-Dispersion
chenliu-1996.github.ioThe official code and project write-up for an ICML 2026 paper showing that smaller language models compress their token embeddings into a narrow region, and proposing a dispersion loss to counteract it.
DeLM: Decentralized Language Models
github.comA research framework that replaces a single central orchestrator with a shared, verified context so parallel agent threads can accumulate reusable progress on a coding benchmark.
DELEGATE-52
arxiv.orgA benchmark and code release for evaluating how LLMs corrupt documents over long-horizon delegated editing, simulating round-trip edits across 52 professional domains such as source code and accounting ledgers, scoring 19 LLMs by document recovery.
Video Compression Commander
xuyang-liu16.github.ioA plug-and-play token-compression framework for video LLMs that dynamically allocates a frame token budget, cutting a model such as LLaVA-OV's generation time by over 70 percent while keeping most accuracy, unifying several published baselines.
A curated, continuously updated list of papers applying large language models to prognostics and health management, covering fault diagnosis and remaining-useful-life prediction, each entry giving the English and Chinese titles, year, venue and code link where available.
CrossCoder
github.comA research framework for repository-level code generation that retrieves context from external libraries as well as the local repository, using a knowledge graph over both.
Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
x-zheng16.github.ioA survey and curated catalog of more than 500 papers on safety risks, attacks and defenses in embodied AI, spanning perception, cognition, planning, interaction and agentic systems.
ICLR2026-Guide-CN
github.comA static, zero-backend web site giving Chinese-language, six-dimension summaries of all 5,352 accepted ICLR 2026 papers, browsable by a two-level topic hierarchy, with the full generation pipeline included.
Awesome Process Reward Models
github.comA comprehensive, continually updated collection of process reward models, models that score each reasoning step rather than only the final answer, for mathematical and other LLM tasks.