MLSys Radar

research

Anthropic · model-lab · 2026-06-03

Research

Score 14

No feed summary available yet.

research

Open

High signal Matched: research

LG AI Research · korea · 2026-06-03

LG AI Research

Score 11

No feed summary available yet.

research

Open

High signal Matched: research

Nota AI · korea · 2026-05-29

Full-Stack Optimization for Low-Light Video on Jetson Orin NX: From 400 ms to 28 ms

Score 23

  Jaehoon Lee Technical Content Manager, Nota AI   When enterprises adopt AI, the most common bottleneck is not model development. It is the deployment stage: getting a finished model to run reliably on the actual target device.T...

inference serving benchmark hardware model-release research quantization evals

Open

High signal Matched: inference, throughput, benchmark, performance, latency, cost, gpu, model, evaluation, quantization, int8, benchmarks, leaderboard

AWS Machine Learning Blog · cloud · 2026-05-29

Evaluating Deep Agents using LangSmith on AWS

Score 9

This post combines learnings from LangChain’s work on evaluating deep agents and Anthropic’s guide to demystifying evals for AI agents into a practical guide. In this post, you will learn how to: 1) apply five evaluation patterns for deep...

research cloud evals agents

Open

High signal Matched: evaluation, bedrock, evals, evaluating, agent, agents

AWS Machine Learning Blog · cloud · 2026-05-29

Build a test suite that grows with your agent with dataset management in Amazon Bedrock AgentCore

Score 13

Datasets in AgentCore is in public preview. Agent evaluation is most powerful when you combine fast-moving online signals with stable offline baselines. To understand whether your agent is truly improving over time, you need a fixed benchm...

benchmark research cloud evals agents

Open

High signal Matched: benchmark, evaluation, bedrock, agent

Nota AI · korea · 2026-05-11

[NetsPresso® x AI Agents] Easier to Use, Even More Powerful

Score 52

  Jaehoon Lee Technical Content Manager, Nota AI   NetsPresso® now embraces AI agents. An easy-to-use interface sits on top of the validated pipeline that handles everything from model compression to device deployment.When a user...

inference serving kernel speculative-decoding moe benchmark hardware model-release research quantization evals agents api

Open

High signal Matched: inference, endpoint, kernel, verification, moe, benchmark, latency, cost, gpu, release, model, evaluation, quantization, quantized, int4, evaluate, benchmarks, swe-bench, mmlu, agent, agents, api

BAIR · research · 2026-05-08

Adaptive Parallel Reasoning: The Next Paradigm in Efficient Inference Scaling

Score 28

.apr-fig { text-align: center; margin: 1.35em 0; line-height: 1.4; } .apr-fig--wide img { display: inline-block; width: 100%; max-width: 100%; height: auto; vertical-align: middle; } .apr-fig--wide-0-8 { max-width: 80%; margin-left: auto;...

inference serving kv-cache speculative-decoding benchmark model-release research training fine-tuning evals long-context agents frontier-model

Open

High signal Matched: inference, decoding, prefill, generation, serve, throughput, kv cache, verification, performance, latency, cost, model, paper, research, evaluation, training, pretraining, sft, benchmarks, long context, context window, agentic, reasoning model

Nota AI · korea · 2026-04-29

[NVIDIA Nemotron Hackathon] Grand Prize Among 20 Teams: Behind Two Sleepless Days

Score 32

  Hancheol Park, Ph. D.AI Research Engineer, NetsPresso Tech, Nota AI Geonmin Kim, Ph. D.AI Research Engineer, NetsPresso Tech, Nota AI Geonho LeeEdge AI Engineer Intern, NetsPresso Tech, Nota AI Jaehoon Lee Technical Content Manager,...

inference moe benchmark model-release research korea training fine-tuning quantization evals agents

Open

High signal Matched: generation, moe, performance, model, weights, paper, research, evaluation, korea, korean, seoul, naver, training, fine-tuning, quantization, agent, agents, agentic

Together AI · inference-infra · 2026-04-29

DeepSeek-V4 Pro now available on Together AI

Score 10

DeepSeek-V4 Pro is now available on Together AI with 512K context, controllable reasoning modes, and cached-input pricing for long-context reasoning workloads like code agents, document intelligence, and research synthesis.

research long-context agents

Open

High signal Matched: research, long-context, agents

Nota AI · korea · 2026-04-22

[Deep Dive: NetsPresso®] From Quantization to Graph Optimization: A Step-by-Step Model Deployment Pipeline

Score 54

  Jaehoon Lee Technical Content Manager, Nota AI   Series Notice: NetsPresso® Technical Blog, Part 2In Part 1, we walked through a scenario of deploying Llama 3.2 1B on an edge device to illustrate the NetsPresso® workflow. The f...

inference kernel cuda benchmark hardware model-release research korea training quantization evals api open-source

Open

High signal Matched: inference, kernel, cuda, matmul, benchmark, performance, latency, cost, npu, model, weights, paper, research, evaluation, furiosa, training, quantization, int8, int4, awq, gptq, sdk, open-source

Nota AI · korea · 2026-04-08

[Overview: NetsPresso®] A Platform That Handles Everything from Model Optimization to Target Deployment

Score 36

  Jaehoon Lee Technical Content Manager, Nota AI   AI Model Optimization: Why Models Won't Run on HardwareThe Chip Is Ready, but the Model Won't DeployIf you have ever tried deploying an AI model onto your own chip, the following...

inference distributed kv-cache speculative-decoding benchmark hardware model-release research quantization evals

Open

High signal Matched: inference, multi-gpu, kv cache, verification, performance, latency, gpu, model, research, evaluation, quantization, quantized, awq, gptq, evaluate

Nota AI · korea · 2026-03-31

The Real Reason TurboQuant Shook the Market: AI Optimization Has Gone Mainstream

Score 46

  Jaehoon Lee Technical Content Manager, Nota AI   In March, a single official announcement from Google Research rocked trillions of won in the market capitalization of U.S. infrastructure and semiconductor stocks. The catalyst:...

inference serving kv-cache benchmark hardware model-release research training fine-tuning quantization agents frontier-model

Open

High signal Matched: inference, serving, generation, throughput, kv cache, benchmark, performance, cost, b200, blackwell, introducing, model, fp8, research, training, fine-tuning, quantization, quantized, agent, agentic, frontier model

Nota AI · korea · 2026-03-23

[GTC 2026 Recap] The Trillion-Dollar Inference Race Begins: How Nota AI Fills the Gap

Score 42

  Jaehoon Lee Technical Content Manager, Nota AI   GTC has evolved far beyond a technology conference, drawing attention from global economies and financial markets alike. This year, CEO Jensen Huang took the stage in his tradema...

inference serving kernel cuda kv-cache benchmark hardware model-release research cloud training long-context agents open-source

Open

High signal Matched: inference, prefill, generation, throughput, cuda, kv cache, performance, latency, cost, gpu, npu, launch, model, research, cloud, training, long-context, context window, agent, agents, agentic, open-source

Nota AI · korea · 2026-03-20

GenAI Everywhere: The Future of Edge AI Optimization with the New NetsPresso®

Score 26

  NP Product Team, Nota AI   The role of Edge AI is rapidly expanding.Offline voice assistants now carry on conversations in our daily lives, vehicles infer routes in real time, and smartphones generate images without a network c...

inference kv-cache moe benchmark model-release research korea quantization

Open

High signal Matched: inference, kv cache, moe, benchmark, performance, latency, cost, model, research, seoul, quantization

Nota AI · korea · 2026-03-13

NotaMoEQuantization: An MoE-Specific Quantization Method for Solar-Open-100B

Score 62

  Hancheol Park, Ph. D. AI Research Engineer, Nota AI Tairen PiaoAI Research Engineer, Nota AI Tae-Ho KimCTO & Co-Founder, Nota AI ✔️ Resource : The official quantized model of Solar-Open-100B, which passed the first round of Sout...

inference serving moe benchmark hardware model-release research korea training quantization evals long-context open-source

Open

High signal Matched: inference, serving, prefill, generation, throughput, moe, router, benchmark, performance, latency, ttft, tpot, blackwell, release, model, weights, open model, research, evaluation, korea, korean, upstage, training, post-training, quantization, quantized, int4, evaluate, benchmarks, mmlu, long-context

BAIR · research · 2026-03-13

Identifying Interactions at Scale for LLMs

Score 18

--> Understanding the behavior of complex machine learning systems, particularly Large Language Models (LLMs), is a critical challenge in modern artificial intelligence. Interpretability research aims to make the decision-making process mo...

inference serving benchmark model-release research training evals long-context rag

Open

High signal Matched: inference, serving, decoding, performance, cost, model, research, training, evaluate, mmlu, long-context, rag

vLLM Project · open-source · 2026-03-04

vLLM Triton Attention Backend Deep Dive

Score 14

This article is adapted from a Red Hat hosted vLLM Office Hours session with Burkhard Ringlein from IBM Research, featuring a deep technical walkthrough of the vLLM Triton attention backend....

kernel triton research

Open

High signal Matched: triton, research

Together AI · inference-infra · 2026-03-02

Introducing Together AI’s new look

Score 14

We've refreshed our visual identity — designed with Pentagram to express how Together AI connects open-source innovation, systems research, and builders to unlock new possibilities.

model-release research open-source

Open

High signal Matched: introducing, research, open-source

Nota AI · korea · 2026-02-26

ERGO: Efficient High-Resolution Visual Understanding for Vision-Language Models

Score 24

  Jewon Lee | Wooksu Shin | Seungmin Yang | Ki-Ung Song | Donguk Lim | Jaeyeon Kim | Tae-Ho Kim |  Bo-Kyeong KimEdgeFM Team, Nota AI ✔️ Resources for more information: GitHub, ArXiv, Project Page, Demo.✔️ Accepted at ICLR 2026. &...

inference speculative-decoding benchmark model-release research training evals

Open

High signal Matched: inference, generation, verification, benchmark, performance, latency, cost, model, arxiv, evaluation, training, post-training, benchmarks

Together AI · inference-infra · 2026-02-06

What do LLMs think when you don't tell them what to think about?

Score 10

What do language models generate when you don't tell them what to generate? New research reveals that LLM families have distinct 'knowledge priors'—GPT models default to code and math, Llama favors narratives, DeepSeek generates religious...

research

Open

High signal Matched: research

Together AI · inference-infra · 2026-01-26

DSGym: A holistic framework for evaluating and training data science agents

Score 18

Introducing DSGym—a holisti evaluation and training framework for LLM-based data science agents. Features 90+ bioinformatics tasks, 92 Kaggle competitions, and synthetic trajectory generation. Our 4B model achieves state-of-the-art perform...

inference benchmark model-release research training evals agents open-source

Open

High signal Matched: generation, performance, introducing, model, evaluation, training, evaluating, agents, open-source

BAIR · research · 2026-01-10

Information-Driven Design of Imaging Systems

Score 12

An encoder (optical system) maps objects to noiseless images, which noise corrupts into measurements. Our information estimator uses only these noisy measurements and a noise model to quantify how well measurements distinguish objects. Man...

benchmark model-release research training evals

Open

High signal Matched: performance, model, paper, evaluation, training, evaluate

Nota AI · korea · 2025-12-19

NVIDIA Blackwell; The Impact of NVFP4 For LLM Inference

Score 74

  Seungmin YangEdgeFM Lead, Nota AI On this page ▾ SummaryWith the introduction of NVFP4—a new 4-bit floating point data type in NVIDIA’s Blackwell GPU architecture—LLM inference achieves markedly improved efficiency.Blackwell’s NVFP4...

inference serving kernel cuda distributed benchmark hardware model-release research training quantization evals rag

Open

High signal Matched: inference, serving, decoding, prefill, generation, token generation, throughput, kernel, gemm, cutlass, distributed, benchmark, performance, latency, ttft, tpot, tokens/sec, cost, gpu, blackwell, launch, model, weights, fp8, research, training, post-training, quantization, quantized, awq, benchmarks, mmlu, retrieval

Together AI · inference-infra · 2025-12-17

Research POV: Yes, AGI Can Happen – A Computational Perspective

Score 14

Dan Fu, our VP of Kernels, has published a new post challenging the idea that AI is hitting a hardware wall. He argues that we are vastly underutilizing current chips and that better software-hardware co-design will unlock the next order o...

benchmark research

Open

High signal Matched: performance, research

AIBrix · open-source · 2025-11-10

AIBrix v0.5.0 Release: Batch API, KVCache v1 Connector, and Enhanced P/D orchestration

Score 22

🚀 AIBrix v0.5.0 Release Today, we’re excited to announce AIBrix v0.5.0, a release that pushes AIBrix closer to a batteries-included control plane for modern LLM workloads. This release introduces an OpenAI-compatible Batch API for hi...

inference benchmark model-release research evals api

Open

High signal Matched: prefill, latency, release, evaluation, api, openai-compatible

BAIR · research · 2025-09-01

What exactly does word2vec learn?

Score 14

What exactly does word2vec learn, and how? Answering this question amounts to understanding representation learning in a minimal yet interesting language modeling task. Despite the fact that word2vec is a well-known precursor to modern lan...

benchmark model-release research training

Open

High signal Matched: benchmark, performance, model, weights, paper, training

Nota AI · korea · 2025-07-10

Video Self-Distillation for Single-Image Encoders: Learning Temporal Priors from Unlabeled Video

Score 20

  Marcel Simon, Ph. D.ML Researcher, Nota AI GmbH Tae-Ho KimCTO & Co-Founder, Nota AI Seul-Ki Yeom, Ph. D.Research Lead, Nota AI GmbH   SummaryProposes a simple next-frame prediction task using unlabeled video to enhance sing...

inference benchmark model-release research training fine-tuning evals

Open

High signal Matched: inference, performance, model, paper, research, training, fine-tuning, benchmarks

BAIR · research · 2025-07-01

Whole-Body Conditioned Egocentric Video Prediction

Score 10

.modal { display: none; position: fixed; z-index: 9999; padding-top: 50px; left: 0; top: 0; width: 100%; height: 100%; overflow: auto; background-color: rgba(0,0,0,0.9); } .modal-content { margin: auto; display: block; max-width: 90%; max-...

inference benchmark model-release research training evals agents

Open

High signal Matched: inference, generation, performance, model, paper, arxiv, evaluation, training, evaluate, agent, agents

Nota AI · korea · 2025-05-08

SplitQuant: Layer Splitting for Low-Bit Neural Network Quantization for Edge AI Devices

Score 20

  Jaewoo SongSoftware Engineer, Nota AI   SummaryThis study proposes an AI model preprocessing method for improved quantization accuracies on edge AI devices which do not support advanced quantization methods due to their limitat...

benchmark model-release research quantization

Open

High signal Matched: performance, model, weights, research, quantization, int8, int4

Nota AI · korea · 2025-05-07

Efficient LLaMA-3.2-Vision by Trimming Cross-attended Visual Features</span#x3E;

Score 28

&nbsp; Jewon Lee | Ki-Ung Song | Seungmin Yang | Donguk Lim | Jaeyeon Kim | Wooksu Shin | Bo-Kyeong Kim | Tae-Ho KimEdgeFM Team, Nota AI Yong Jae Lee, Ph. D.Associate Professor, UW-Madison &nbsp; SummaryOur method, Trimmed-Llama, reduces t...

inference kv-cache benchmark model-release research training evals open-source

Open

High signal Matched: inference, generation, kv cache, benchmark, performance, latency, model, weights, research, training, benchmarks, open-source

Modal · inference-infra · 2025-04-18

How sync. uses Modal to lipsync 100 hours of video a day

Score 8

sync. is a research lab training foundational models to understand and manipulate humans in video. After outgrowing Google Colab, they partnered with Modal for efficient deployment, allowing rapid iteration and scaling to process over 100...

research training

Open

High signal Matched: research, training

BAIR · research · 2025-04-11

Defending against Prompt Injection with Structured Queries (StruQ) and Preference Optimization (SecAlign)

Score 10

Recent advances in Large Language Models (LLMs) enable exciting LLM-integrated applications. However, as LLMs have improved, so have the attacks against them. Prompt injection attack is listed as the #1 threat by OWASP to LLM-integrated ap...

benchmark model-release research training fine-tuning evals rag api frontier-model

Open

High signal Matched: cost, model, evaluation, training, dpo, fine-tuning, retrieval, api, sota

BAIR · research · 2025-04-08

Repurposing Protein Folding Models for Generation with Latent Diffusion

Score 20

PLAID is a multimodal generative model that simultaneously generates protein 1D sequence and 3D structure, by learning the latent space of protein folding models. The awarding of the 2024 Nobel Prize to AlphaFold2 marks an important moment...

inference benchmark model-release research training rag

Open

High signal Matched: inference, generation, cost, model, weights, research, training, retrieval

Nota AI · korea · 2025-04-08

UniForm: A Reuse Attention Mechanism for Efficient Transformers on Resource-Constrained Edge Devices

Score 24

&nbsp; Seul-Ki Yeom, Ph. D. Research Lead, Nota AI GmbH Tae-Ho KimCTO &amp; Co-Founder, Nota AI &nbsp; SummaryDelivers real-time AI performance on edge devices such as smartphones, IoT devices, and embedded systems.Introduces a novel "Reus...

inference kernel benchmark model-release research evals

Open

High signal Matched: inference, kernel, benchmark, performance, cost, introducing, model, paper, research, benchmarks

SkyPilot · open-source · 2025-02-26

Using DeepSeek R1 for RAG: Do's and Don'ts

Score 10

DeepSeek R1 has shown great reasoning capability when it is firstly released. In this blog post, we detail our learnings in using DeepSeek R1 to build a Retrieval-Augmented Generation (RAG) system, tailored for legal documents. We choose l...

inference research rag

Open

High signal Matched: generation, research, rag, retrieval-augmented generation, retrieval

SqueezeBits · korea · 2024-10-01

[vLLM vs TensorRT-LLM] #1. An Overall Evaluation

Score 22

This article provides a comparative analysis of vLLM and TensorRT-LLM frameworks for serving LLMs, evaluating their performance based on key metrics like throughput, TTFT, and TPOT to offer insights for practitioners in optimizing LLM depl...

inference serving benchmark research evals

Open

High signal Matched: serving, throughput, performance, ttft, tpot, evaluation, evaluating

Nota AI · korea · 2024-08-02

Deploying an Efficient Vision-Language Model on Mobile Devices

Score 38

&nbsp; Jaeyeon KimResearch Engineer, Nota AI Geonmin KimResearch Engineer, Nota AI Hancheol ParkTeam Lead of NetsPresso Application, Nota AI &nbsp; IntroductionRecent large language models (LLMs) have demonstrated unprecedented performance...

inference benchmark model-release research cloud training fine-tuning evals open-source

Open

High signal Matched: decoding, benchmark, performance, latency, tokens/sec, model, arxiv, research, technical report, evaluation, cloud, training, lora, benchmarks, leaderboard, open-source

Microsoft Research · big-tech · 2026-05-29

Data Formulator 0.7: AI-powered data analytics for enterprise data

Score 5

Data Formulator introduces AI-powered analytics for enterprise data workflows. Data teams can easily bring enterprise data into an AI-ready workspace where users can explore, analyze, and visualize data with AI agents to turn raw data into...

research agents

Open

Watchlist Matched: research, agents

Microsoft Research · big-tech · 2026-05-28

Extending Human Intelligence Through AI

Score 5

Understanding AI as an extension of human intelligence—not a replacement for it—offers a more grounded path for building trustworthy AI systems. The post Extending Human Intelligence Through AI appeared first on Microsoft Research.

research

Open

Watchlist Matched: research

BAIR · research · 2025-11-01

RL without TD learning

Score 4

In this post, I’ll introduce a reinforcement learning (RL) algorithm based on an “alternative” paradigm: divide and conquer. Unlike traditional methods, this algorithm is not based on temporal difference (TD) learning (which has scalabilit...

benchmark model-release research training

Open

Watchlist Matched: benchmark, performance, model, paper, training

BAIR · research · 2025-03-25

Scaling Up Reinforcement Learning for Traffic Smoothing: A 100-AV Highway Deployment

Score 6

Training Diffusion Models with Reinforcement Learning We deployed 100 reinforcement learning (RL)-controlled cars into rush-hour highway traffic to smooth congestion and reduce fuel consumption for everyone. Our goal is to tackle "stop-and...

serving kernel benchmark model-release research training agents

Open

Watchlist Matched: throughput, kernel, performance, model, paper, training, agent, agents