Papersby Glyph

Notable data and AI papers

Papers627 Labs45

627 papers · 45 labs

Mamba: Linear-Time Sequence Modeling with Selective State Spaces research paper by Carnegie Mellon University, 2023

Carnegie Mellon University · Architectures · Dec 1, 2023 · 10,151 citations · 152 upvotes

A selective state space model that scales linearly with sequence length and matches Transformers on language.

By Albert Gu, Tri Dao · arXiv 2312.00752 · PDF · Code

Summarised by hand from the abstract.

More from Carnegie Mellon University

Agents and evaluationUnverifiedOct 31, 2026Random Is Almost Enough? Stable and Low-Cost Benchmark Compression via BestRandBestRand is a method for selecting small benchmark subsets that reliably estimate full evaluation scores at lower computational costs.

Carnegie Mellon University51 upvotesRead →Foundation modelsUnverifiedOct 24, 2026Eat Your Precision: Precision Objective Mitigates Reward Hacking in Rubric-based RLFORK jointly optimizes recall and precision via rubric-driven necessity credits and factual grading to improve response quality and reduce verbosity in reinforcement learning.

Carnegie Mellon University34 upvotesRead →Agents and evaluationUnverifiedOct 13, 2026Into the Unknown: Benchmarking AI agents on open scientific research problemsThis benchmark evaluates how well AI agents can write follow-up scientific research papers based on unpublished directions from domain experts.

Carnegie Mellon University62 upvotesRead →Applied AIUnverifiedOct 12, 2026X-Ragent: Tool-Augmented Reinforcement Learning for Evidence-Grounded Chest X-Ray ReasoningThis paper introduces a reinforcement learning framework for medical agents that selectively uses external evidence to improve diagnostic reasoning in chest X-rays.

Carnegie Mellon University44 upvotesRead →ReasoningUnverifiedOct 1, 2026Selection-Based Structured Reasoning: Toward Efficient Multimodal Search AgentsSSR improves multimodal agent efficiency by replacing free-form reasoning generation with a selection process from pre-specified reasoning candidates.

Carnegie Mellon University37 upvotesRead →Applied AIUnverifiedSep 29, 2026Predictive Credit: Measuring What Scientific Explanations Add to Experimental ForecastsThis paper introduces a protocol to measure how much scientific explanations actually improve the accuracy of experimental forecasts.

Carnegie Mellon University103 upvotesRead →Multimodal and roboticsUnverifiedSep 23, 2026Training Object Permanence in World ModelsThis paper introduces a dataset and training stack to improve object permanence and physical reasoning in video world models.

Carnegie Mellon University242 upvotesRead →Agents and evaluationUnverifiedSep 22, 2026JEV-as-a-Judge: Accept When Confident, Escalate When UnsureJEV-as-a-Judge is a cost-effective evaluation framework that uses a simple decision-only judge for easy cases and escalates complex ones.

Carnegie Mellon University57 citations42 upvotesRead →

ArchitecturesJun 12, 2017Attention Is All You NeedIntroduced the Transformer, the attention-only architecture behind nearly every large language model since.

Google201k citations142 upvotesRead →ArchitecturesSep 2, 2021CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and GenerationCodeT5, a unified encoder-decoder Transformer model, improves code understanding and generation by leveraging semantic information from identifiers and user comments.

Salesforce2,633 citations3 upvotesRead →ArchitecturesUnverifiedMar 14, 2025VGGT: Visual Geometry Grounded TransformerVGGT, a feed-forward neural network, efficiently infers multiple 3D attributes from single or multiple views, outperforming alternatives and enhancing downstream tasks without post-processing.

Meta2,132 citations40 upvotesRead →ArchitecturesUnverifiedJan 11, 2024DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language ModelsThe DeepSeekMoE architecture improves expert specialization in Mixture-of-Experts models by segmenting experts and isolating shared ones, achieving better performance and computational efficiency compared to GShard and other models.

DeepSeek1,268 citations65 upvotesRead →ArchitecturesMay 13, 2023CodeT5+: Open Code Large Language Models for Code Understanding and GenerationCodeT5+, a family of flexible encoder-decoder LLMs initialized with frozen LLMs and trained with a mixture of pretraining objectives, achieves state-of-the-art performance across various code-related tasks.

Salesforce836 citations7 upvotesRead →ArchitecturesUnverifiedOct 9, 2019HuggingFace's Transformers: State-of-the-art Natural Language ProcessingTransformers library provides state-of-the-art Transformer architectures and pretrained models for natural language processing tasks with a unified API and emphasis on extensibility and robust deployment.

Hugging Face251 citations34 upvotesRead →

About this paper

Venue: arXiv.org
Citations: 10,151, 1,413 influential · Semantic Scholar
Upvotes: 152 · Hugging Face
Lab: Carnegie Mellon University

Changes

CitationsNov 9, 2026Influential citations rose from 1,397 to 1,413.

CitationsNov 9, 2026Citations rose from 10,017 to 10,151.

CitationsNov 2, 2026Influential citations rose from 1,354 to 1,397.

CitationsNov 2, 2026Citations rose from 9,539 to 10,017.

CitationsOct 12, 2026Influential citations rose from 1,331 to 1,354.

CitationsOct 12, 2026Citations rose from 9,373 to 9,539.

CitationsOct 5, 2026Influential citations rose from 1,319 to 1,331.

CitationsOct 5, 2026Citations rose from 9,252 to 9,373.

CitationsSep 28, 2026Influential citations rose from 1,315 to 1,319.

CitationsSep 28, 2026Citations rose from 9,203 to 9,252.

CitationsSep 25, 2026Influential citations rose from 1,312 to 1,315.

CitationsSep 25, 2026Citations rose from 9,182 to 9,203.

New paperSep 24, 2026Added to the list.