Paper List

2405

PaliGemma

Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis

Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality

Kaleido Diffusion: Improving Conditional Diffusion Models with Autoregressive Latent Modeling

4Diffusion: Multi-view Video Diffusion Model for 4D Generation

Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models

MotionLLM: Understanding Human Behaviors from Human Motions and Videos

Xwin-LM: Strong and Scalable Alignment Practice for LLMs

GECO: Generative Image-to-3D within a SECOnd

DITTO-2: Distilled Diffusion Inference-Time T-Optimization for Music Generation

MOFA-Video: Controllable Image Animation via Generative Motion Field Adaptions in Frozen Image-to-Video Diffusion Model

Jina CLIP: Your CLIP Model Is Also Your Text Retriever

PLA4D: Pixel-Level Alignments for Text-to-4D Gaussian Splatting

Similarity is Not All You Need: Endowing Retrieval Augmented Generation with Multi Layered Thoughts

Parrot: Efficient Serving of LLM-based Applications with Semantic Variable

DevEval: A Manually-Annotated Code Generation Benchmark Aligned with Real-World Code Repositories

DeMamba: AI-Generated Video Detection on Million-Scale GenVideo Benchmark

Self-Exploring Language Models: Active Preference Elicitation for Online Alignment

NPGA: Neural Parametric Gaussian Avatars

MAP-Neo: Highly Capable and Transparent Bilingual Large Language Model Series

Nearest Neighbor Speculative Decoding for LLM Generation and Attribution

Value-Incentivized Preference Optimization: A Unified Approach to Online and Offline RLHF

Offline Regularised Reinforcement Learning for Large Language Models Alignment

EasyAnimate: A High-Performance Long Video Generation Method based on Transformer Architecture

LLMs achieve adult human performance on higher-order theory of mind tasks

T2V-Turbo: Breaking the Quality Bottleneck of Video Consistency Model with Mixed Reward Feedback

Zipper: A Multi-Tower Decoder Architecture for Fusing Modalities

Atlas3D: Physically Constrained Self-Supporting Text-to-3D for Simulation and Fabrication

SoundCTM: Uniting Score-based and Consistency Models for Text-to-Sound Generation

GFlow: Recovering 4D World from Monocular Video

3DitScene: Editing Any Scene via Language-guided Disentangled Gaussian Splatting

Phased Consistency Model

Instruct-MusicGen: Unlocking Text-to-Music Editing for Music Language Models via Instruction Tuning

LLaMA-NAS: Efficient Neural Architecture Search for Large Language Models

4-bit Shampoo for Memory-Efficient Network Training

2BP: 2-Stage Backpropagation

VeLoRA: Memory Efficient Training using Rank-1 Sub-Token Projections

Yuan 2.0-M32: Mixture of Experts with Attention Router

Matryoshka Multimodal Models

Collaborative Video Diffusion: Consistent Multi-video Generation with Camera Control

Human4DiT: Free-view Human Video Generation with 4D Diffusion Transformer

Trans-LoRA: towards data-free Transferable Parameter Efficient Finetuning

An Introduction to Vision-Language Modeling

Part123: Part-aware 3D Reconstruction from a Single-view Image

Greedy Growing Enables High-Resolution Pixel-Based Diffusion Models

ConvLLaVA: Hierarchical Backbones as Visual Encoder for Large Multimodal Models

The Road Less Scheduled

Automatic Data Curation for Self-Supervised Learning: A Clustering-Based Approach

Meteor: Mamba-based Traversal of Rationale for Large Language and Vision Models

Stacking Your Transformers: A Closer Look at Model Growth for Efficient LLM Pre-Training

iVideoGPT: Interactive VideoGPTs are Scalable World Models

Denoising LM: Pushing the Limits of Error Correction Models for Speech Recognition

ODGEN: Domain-specific Object Detection Data Generation with Diffusion Models

OptLLM: Optimal Assignment of Queries to Large Language Models

HDR-GS: Efficient High Dynamic Range Novel View Synthesis at 1000x Speed via Gaussian Splatting

Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization

Aya 23: Open Weight Releases to Further Multilingual Progress

CraftsMan: High-fidelity Mesh Generation with 3D Native Generation and Interactive Geometry Refiner

Data Mixing Made Efficient: A Bivariate Scaling Law for Language Model Pretraining

AutoCoder: Enhancing Code Large Language Model with \textsc{AIEV-Instruct}

NeRF-Casting: Improved View-Dependent Appearance with Consistent Reflections

Improved Distribution Matching Distillation for Fast Image Synthesis

Tele-Aloha: A Low-budget and High-authenticity Telepresence System Using Sparse RGB Cameras

Not All Language Model Features Are Linear

Semantica: An Adaptable Image-Conditioned Diffusion Model

Neural Directional Encoding for Efficient and Accurate View-Dependent Appearance Modeling

RectifID: Personalizing Rectified Flow with Anchored Classifier Guidance

Visual Echoes: A Simple Unified Transformer for Audio-Visual Generation

LiteVAE: Lightweight and Efficient Variational Autoencoders for Latent Diffusion Models

DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data

DiM: Diffusion Mamba for Efficient High-Resolution Image Synthesis

AlignGPT: Multi-modal Large Language Models with Adaptive Alignment Capability

Distributed Speculative Inference of Large Language Models

Affine-based Deformable Attention and Selective Fusion for Semi-dense Matching

ReVideo: Remake a Video with Motion and Content Control

Thermodynamic Natural Gradient Descent

Dense Connector for MLLMs

CamViG: Camera Aware Image-to-Video Generation with Multimodal Transformers

KPConvX: Modernizing Kernel Point Convolution with Kernel Attention

Reducing Transformer Key-Value Cache Size with Cross-Layer Attention

OmniGlue: Generalizable Feature Matching with Foundation Model Guidance

Personalized Residuals for Concept-Driven Text-to-Image Generation

Face Adapter for Pre-Trained Diffusion Models with Fine-Grained ID and Attribute Control

Retrieval-Augmented Language Model for Extreme Multi-Label Knowledge Graph Link Prediction

Quantifying Emergence in Large Language Models

Diffusion for World Modeling: Visual Details Matter in Atari

Your Transformer is Secretly Linear

Octo: An Open-Source Generalist Robot Policy

Training Data Attribution via Approximate Unrolled Differentiation

MoRA: High-Rank Updating for Parameter-Efficient Fine-Tuning

Imp: Highly Capable Large Multimodal Models for Mobile Devices

On Efficient and Statistical Quality Estimation for Data Annotation

Information Leakage from Embedding in Large Language Models

SLAB: Efficient Transformers with Simplified Linear Attention and Progressive Re-parameterized Batch Normalization

FIFO-Diffusion: Generating Infinite Videos from Text without Training

Dreamer XL: Towards High-Resolution Text-to-3D Generation via Trajectory Score Matching

Towards Modular LLMs by Building and Reusing a Library of LoRAs

OpenRLHF: An Easy-to-use, Scalable and High-performance RLHF Framework

Observational Scaling Laws and the Predictability of Language Model Performance

INDUS: Effective and Efficient Language Models for Scientific Applications

Layer-Condensed KV Cache for Efficient Inference of Large Language Models

Dynamic data sampler for cross-language transfer learning in large language models

Grounded 3D-LLM with Referent Tokens

Toon3D: Seeing Cartoons from a New Perspective

TRANSIC: Sim-to-Real Policy Transfer by Learning from Online Correction

CAT3D: Create Anything in 3D with Multi-View Diffusion Models

Grounding DINO 1.5: Advance the "Edge" of Open-Set Object Detection

Dual3D: Efficient and Consistent Text-to-3D Generation with Dual-mode Multi-view Latent Diffusion

Chameleon: Mixed-Modal Early-Fusion Foundation Models

Many-Shot In-Context Learning in Multimodal Foundation Models

LoRA Learns Less and Forgets Less

BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation

ALPINE: Unveiling the Planning Capability of Autoregressive Learning in Language Models

Xmodel-VLM: A Simple Baseline for Multimodal Vision Language Model

Naturalistic Music Decoding from EEG Data via Latent Diffusion Models

Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding

Beyond Scaling Laws: Understanding Transformer Performance with Associative Memory

Understanding the performance gap between online and offline alignment algorithms

No Time to Waste: Squeeze Time into Channel for Mobile Video Understanding

SpeechGuard: Exploring the Adversarial Robustness of Multimodal Large Language Models

SpeechVerse: A Large-scale Generalizable Audio Language Model

Compositional Text-to-Image Generation with Dense Blob Representations

Coin3D: Controllable and Interactive 3D Assets Generation with Proxy-Guided Conditioning

Plot2Code: A Comprehensive Benchmark for Evaluating Multi-modal Large Language Models in Code Generation from Scientific Plots

The Platonic Representation Hypothesis

PARDEN, Can You Repeat That? Defending against Jailbreaks via Repetition

Zero-Shot Tokenizer Transfer

RLHF Workflow: From Reward Modeling to Online RLHF

MS MARCO Web Search: a Large-scale Information-rich Web Dataset with Millions of Real Click Labels

SambaNova SN40L: Scaling the AI Memory Wall with Dataflow and Composition of Experts

PromptLink: Leveraging Large Language Models for Cross-Source Biomedical Concept Linking

LogoMotion: Visually Grounded Code Generation for Content-Aware Animation

Piccolo2: General Text Embedding with Multi-task Hybrid Loss Training

SUTRA: Scalable Multilingual Language Model Architecture

Large Language Models as Planning Domain Generators

Linearizing Large Language Models

Mitigating Hallucinations in Large Language Models via Self-Refinement-Enhanced Knowledge Retrieval

A Survey on RAG Meets LLMs: Towards Retrieval-Augmented Large Language Models

Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?

KV-Runahead: Scalable Causal LLM Inference by Parallel Key-Value Cache Generation

You Only Cache Once: Decoder-Decoder Architectures for Language Models

ChuXin: 1.6B Technical Report

xLSTM: Extended Long Short-Term Memory

DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Granite Code Models: A Family of Open Foundation Models for Code Intelligence

ContextQ: Generated Questions to Support Meaningful Parent-Child Dialogue While Co-Reading

AlphaMath Almost Zero: process Supervision without process

MAmmoTH2: Scaling Instructions from the Web

Is Sora a World Simulator? A Comprehensive Survey on General World Models and Beyond

Lory: Fully Differentiable Mixture-of-Experts for Autoregressive Language Model Pre-training

Is Flash Attention Stable?

What matters when building vision-language models?

Customizing Text-to-Image Models with a Single Image Pair

Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models

FLAME: Factuality-Aware Alignment for Large Language Models

NeMo-Aligner: Scalable Toolkit for Efficient Model Alignment

WildChat: 1M ChatGPT Interaction Logs in the Wild

StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generation

LLM-AD: Large Language Model based Audio Description System

LoRA Land: 310 Fine-tuned LLMs that Rival GPT-4, A Technical Report

Spectrally Pruned Gaussian Fields with Neural Compensation

Self-Play Preference Optimization for Language Model Alignment

Is Bigger Edit Batch Size Always Better? -- An Empirical Study on Model Editing with Llama-3

BiomedRAG: A Retrieval Augmented Large Language Model for Biomedicine

A Careful Examination of Large Language Model Performance on Grade School Arithmetic

Clover: Regressive Lightweight Speculative Decoding with Sequential Knowledge

STT: Stateful Tracking with Transformers for Autonomous Driving

SemantiCodec: An Ultra Low Bitrate Semantic Audio Codec for General Sound

Constrained Decoding for Secure Code Generation

A Primer on the Inner Workings of Transformer-based Language Models

SPAFIT: Stratified Progressive Adaptation Fine-tuning for Pre-trained Large Language Models

In-Context Learning with Long-Context Models: An In-Depth Exploration

Automatic Creative Selection with Cross-Modal Matching

2404

OpenEQA: Embodied Question Answering in the Era of Foundation Models

CodeGemma: Open Code Models Based on Gemma

Lightplane: Highly-Scalable Components for Neural 3D Fields

MotionLCM: Real-time Controllable Motion Generation via Latent Consistency Model

Invisible Stitch: Generating Smooth 3D Scenes with Depth Inpainting

KAN: Kolmogorov-Arnold Networks

DOCCI: Descriptions of Connected and Contrasting Images

Visual Fact Checker: Enabling High-Fidelity Detailed Caption Generation

Better & Faster Large Language Models via Multi-token Prediction

Iterative Reasoning Preference Optimization

When to Retrieve: Teaching LLMs to Utilize Information Retrieval Effectively

GS-LRM: Large Reconstruction Model for 3D Gaussian Splatting

Extending Llama-3's Context Ten-Fold Overnight

RAG and RAU: A Survey on Retrieval-Augmented Language Model in Natural Language Processing

MicroDreamer: Zero-shot 3D Generation in $\sim$20 Seconds by Score-based Iterative Reconstruction

InstantFamily: Masked Attention for Zero-shot Multi-ID Image Generation

Octopus v4: Graph of language models

SAGS: Structure-Aware 3D Gaussian Splatting

Hallucination of Multimodal Large Language Models: A Survey

Stylus: Automatic Adapter Selection for Diffusion Models

Kangaroo: Lossless Self-Speculative Decoding via Double Early Exiting

Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models

PECC: Problem Extraction and Coding Challenges

ChatGPT as an inventor: Eliciting the strengths and weaknesses of current large language models against humans in engineering design

Capabilities of Gemini Models in Medicine

LEGENT: Open Platform for Embodied Agents

Paint by Inpaint: Learning to Add Image Objects by Removing Them First

BlenderAlchemy: Editing 3D Graphics with Vision-Language Models

MaPa: Text-driven Photorealistic Material Painting for 3D Shapes

Ag2Manip: Learning Novel Manipulation Skills with Agent-Agnostic Visual and Action Representations

Reinforcement Retrieval Leveraging Fine-grained Feedback for Fact Checking News Claims with Black-Box LLM

Small Language Models Need Strong Verifiers to Self-Correct Reasoning

Automated Data Visualization from Natural Language via Large Language Models: An Exploratory Study

PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning

AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs

HaLo-NeRF: Learning Geometry-Guided Semantics for Exploring Unconstrained Photo Collections

How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites

Revisiting Text-to-Image Evaluation with Gecko: On Metrics, Prompts, and Human Ratings

Make Your LLM Fully Utilize the Context

SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension

ConsistentID: Portrait Generation with Multimodal Fine-Grained Identity Preserving

Layer Skip: Enabling Early Exit Inference and Self-Speculative Decoding

Tele-FLM Technical Report

TinyChart: Efficient Chart Understanding with Visual Token Merging and Program-of-Thoughts Learning

Interactive3D: Create What You Want by Interactive 3D Generation

List Items One by One: A New Data Source and Learning Paradigm for Multimodal LLMs

NeRF-XL: Scaling NeRFs with Multiple GPUs

MaGGIe: Masked Guided Gradual Human Instance Matting

MoDE: CLIP Data Experts via Clustering

Editable Image Elements for Controllable Synthesis

PuLID: Pure and Lightning ID Customization via Contrastive Alignment

Leveraging Large Language Models for Multimodal Search

MotionMaster: Training-free Camera Motion Transfer For Video Generation

BASS: Batched Attention-optimized Speculative Sampling

Let's Think Dot by Dot: Hidden Computation in Transformer Language Models

CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data

ID-Aligner: Enhancing Identity-Preserving Text-to-Image Generation with Reward Feedback Learning

XC-Cache: Cross-Attending to Cached Context for Efficient LLM Inference

Label-Efficient Sleep Staging Using Transformers Pre-trained with Position Prediction

Multi-Head Mixture-of-Experts

Transformers Can Represent n-gram Language Models

FlashSpeech: Efficient Zero-Shot Speech Synthesis

Pegasus-1 Technical Report

OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework

Align Your Steps: Optimizing Sampling Schedules in Diffusion Models

SnapKV : LLM Knows What You are Looking for Before Generation

Learning H-Infinity Locomotion Control

SEED-X: Multimodal Models with Unified Multi-granularity Comprehension and Generation

A Multimodal Automated Interpretability Agent

Better Synthetic Data by Retrieving and Transforming Existing Datasets

Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizer

MultiBooth: Towards Generating All Your Concepts in an Image from Text

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

How Good Are Low-bit Quantized LLAMA3 Models? An Empirical Study

Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image Synthesis

Music Consistency Models

The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions

FlowMind: Automatic Workflow Generation with LLMs

PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation

Stronger Random Baselines for In-Context Learning

Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models

Towards Reliable Latent Knowledge Estimation in LLMs: In-Context Learning vs. Prompting Based Factual Knowledge Extraction

LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency

How Far Can We Go with Practical Function-Level Program Repair?

TextSquare: Scaling up Text-Centric Visual Instruction Tuning

AutoCrawler: A Progressive Understanding Web Agent for Web Crawler Generation

Does Gaussian Splatting need SFM Initialization?

HalluciBot: Is There No Such Thing as a Bad Question?

BLINK: Multimodal Large Language Models Can See but Not Perceive

Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models

MeshLRM: Large Reconstruction Model for High-Quality Mesh

From r to Q*: Your Language Model is Secretly a Q-Function

AniClipart: Clipart Animation with Text-to-Video Priors

Reuse Your Rewards: Reward Model Transfer for Zero-Shot Cross-Lingual Alignment

Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing

Introducing v0.5 of the AI Safety Benchmark from MLCommons

OpenBezoar: Small, Cost-Effective and Open Models Trained on Mixes of Instruction Data

EdgeFusion: On-Device Text-to-Image Generation

TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding

Dynamic Typography: Bringing Words to Life

The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey

MoA: Mixture-of-Attention for Subject-Context Disentanglement in Personalized Image Generation

Octopus v3: Technical Report for On-device Sub-billion Multimodal AI Agent

Many-Shot In-Context Learning

A Survey on Retrieval-Augmented Text Generation for Large Language Models

HumMUSS: Human Motion Understanding using State Space Models

Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study

VASA-1: Lifelike Audio-Driven Talking Faces Generated in Real Time

Hierarchical Context Merging: Better Long Context Understanding for Pre-trained LLMs

Social Choice for AI Alignment: Dealing with Diverse Human Feedback

How faithful are RAG models? Quantifying the tug-of-war between RAG and LLMs’ internal prior

Scaling Instructable Agents Across Many Simulated Worlds

Chinchilla Scaling: A replication attempt

Taming Latent Diffusion Model for Neural Radiance Field Inpainting

MMInA: Benchmarking Multihop Multimodal Internet Agents

HQ-Edit: A High-Quality Dataset for Instruction-based Image Editing

CTRL-Adapter: An Efficient and Versatile Framework for Adapting Diverse Controls to Any Diffusion Model

Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization

Compression Represents Intelligence Linearly

Video2Game: Real-time, Interactive, Realistic and Browser-Compatible Environment from a Single Video

Learn Your Reference Model for Real Good Alignment

State Space Model for New-Generation Network Alternative to Transformers: A Survey

CompGS: Efficient 3D Scene Representation via Compressed Gaussian Splatting

TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models

TransformerFAM: Feedback attention is working memory

LLM In-Context Recall is Prompt Dependent

On Speculative Decoding for Multimodal Large Language Models

The Illusion of State in State-Space Models

Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length

JailbreakLens: Visual Analysis of Jailbreak Attacks Against Large Language Models

COCONut: Modernizing COCO Segmentation

Probing the 3D Awareness of Visual Foundation Models

Pre-training Small Base LMs with Fewer Tokens

On the Robustness of Language Guidance for Low-Level Vision Tasks: Findings from Depth Estimation

Dataset Reset Policy Optimization for RLHF

MonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance

Scaling (Down) CLIP: A Comprehensive Analysis of Data, Architecture, and Training Strategies

Reducing hallucination in structured outputs via Retrieval-Augmented Generation

Conformal Prediction via Regression-as-Classification

ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback

LLoCO: Learning Long Contexts Offline

Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models

OSWORLD: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

RHO-1: Not All Tokens Are What You Need

HGRN2: Gated Linear RNNs with State Expansion

RecurrentGemma: Moving Past Transformers for Efficient Open Language Models

Sparse Laneformer

Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models

Audio Dialogues: Dialogues dataset for audio and music understanding

From Words to Numbers: Your Large Language Model Is Secretly A Capable Regressor When Given In-Context Examples

Best Practices and Lessons Learned on Synthetic Data for Language Models

Transferable and Principled Efficiency for Open-Vocabulary Segmentation

JetMoE: Reaching Llama2 Performance with 0.1M Dollars

BRAVE: Broadening the visual encoding of vision-language models

RealmDreamer: Text-Driven 3D Scene Generation with Inpainting and Depth Diffusion

Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention

DreamScene360: Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting

Urban Architect: Steerable 3D Urban Scene Generation with Layout Prior

Adapting LLaMA Decoder to Vision Transformer

RULER: What's the Real Context Size of Your Long-Context Language Models?

InternLM-XComposer2-4KHD: A Pioneering Large Vision-Language Model Handling Resolutions from 336 Pixels to 4K HD

Reconstructing Hand-Held Objects in 3D

pfl-research: simulation framework for accelerating research in Private Federated Learning

Magic-Boost: Boost 3D Generation with Mutli-View Conditioned Diffusion

MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies

MuPT: A Generative Symbolic Music Pretrained Transformer

OmniFusion Technical Report

Elephants Never Forget: Memorization and Learning of Tabular Data in Large Language Models

Revising Densification in Gaussian Splatting

Hash3D: Training-free Acceleration for 3D Generation

Privacy Preserving Prompt Engineering: A Survey

THOUGHTSCULPT: Reasoning with Intermediate Revision and Search

LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders

WILBUR: Adaptive In-Context Learning for Robust and Accurate Web Agents

Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence

CodecLM: Aligning Language Models with Tailored Synthetic Data

SambaLingo: Teaching Large Language Models New Languages

MA-LMM: Memory-Augmented Large Multimodal Model for Long-Term Video Understanding

Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs

SwapAnything: Enabling Arbitrary Object Swapping in Personalized Visual Editing

Evaluating Mathematical Reasoning Beyond Accuracy

MoMA: Multimodal LLM Adapter for Fast Personalized Image Generation

YaART: Yet Another ART Rendering Technology

UniFL: Improve Stable Diffusion via Unified Feedback Learning

Physics of Language Models: Part 3.3, Knowledge Capacity Scaling Laws

MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators

Multilingual Large Language Model: A Survey of Resources, Taxonomy and Frontiers

ByteEdit: Boost, Comply and Accelerate Generative Image Editing

BeyondScene: Higher-Resolution Human-Centric Scene Generation With Pretrained Diffusion

DATENeRF: Depth-Aware Text-based Editing of NeRFs

Q-PEFT: Query-dependent Parameter Efficient Fine-tuning for Text Reranking with Large Language Models

Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Models

Aligning Diffusion Models by Optimizing Human Utility

PhysAvatar: Learning the Physics of Dressed 3D Avatars from Visual Observations

Koala: Key frame-conditioned long video-LLM

SpatialTracker: Tracking Any 2D Pixels in 3D Space

Sigma : Siamese Mamba Network for Multi-Modal Semantic Segmentation

Robust Gaussian Splatting

Social Skill Training with Large Language Models

Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model

No “Zero-Shot” Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance

BuDDIE: A Business Document Dataset for Multi-task Information Extraction

Verifiable by Design: Aligning Language Models to Quote from Pre-Training Data

CantTalkAboutThis: Aligning Language Models to Stay on Topic in Dialogues

Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences

Stream of Search (SoS): Learning to Search in Language

RL for Consistency Models: Faster Reward Guided Text-to-Image Generation

CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching

AutoWebGLM: Bootstrap And Reinforce A Large Language Model-based Web Navigating Agent

Training LLMs over Neurally Compressed Text

ReFT: Representation Finetuning for Language Models

PointInfinity: Resolution-Invariant Point Diffusion Models

CodeEditorBench: Evaluating Code Editing Capability of Large Language Models

Can Small Language Models Help Large Language Models Reason Better?: LM-Guided Chain-of-Thought

MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens

Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?

Scaling Up Video Summarization Pretraining with Large Language Models

RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis

LVLM-Intrepret: An Interpretability Tool for Large Vision-Language Models

Talaria: Interactively Optimizing Machine Learning Models for Efficient Inference

PiSSA: Principal Singular Values and Singular Vectors Adaptation of Large Language Models

Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction

ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline

On the Scalability of Diffusion-based Text-to-Image Generation

Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models

Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Advancing LLM Reasoning Generalists with Preference Trees

Long-context LLMs Struggle with Long In-context Learning

HyperCLOVA X Technical Report

Poro 34B and the Blessing of Multilinguality

Octopus v2: On-device language model for super agent

LLM-ABR: Designing Adaptive Bitrate Algorithms via Large Language Models

Are large language models superhuman chemists?

LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model

HairFastGAN: Realistic and Robust Hair Transfer with a Fast Encoder-Based Approach

2403

ReALM: Reference Resolution As Language Modeling

DiJiang: Efficient Large Language Models through Compact Kernelization

Jamba: A Hybrid Transformer-Mamba Language Model

Localizing Paragraph Memorization in Language Models

Model Stock: All we need is just a few fine-tuned models

sDPO: Don’t Use Your Data All at Once

Towards a World-English Language Model for On-Device Virtual Assistants

BLADE: Enhancing Black-box Large Language Models with Small Domain-Specific Models

The Unreasonable Ineffectiveness of the Deeper Layers

Fully-fused Multi-Layer Perceptrons on Intel Data Center GPUs

Arcee’s MergeKit: A Toolkit for Merging Large Language Models

Evolutionary Optimization of Model Merging Recipes

LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression

mPLUG-DocOwl 1.5: Unified Structure Learning for OCR-free Document Understanding

RAFT: Adapting Language Model to Domain Specific RAG

WavCraft: Audio Editing and Generation with Large Language Models

Gemma: Open Models Based on Gemini Research and Technology

Process Modeling With Large Language Models

VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models

Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Poly-View Contrastive Learning

How Far Are We from Intelligent Visual Deductive Reasoning?

Learning to Decode Collaboratively with Multiple Language Models

GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection

LAB: Large-Scale Alignment for ChatBots

CLLMs: Consistency Large Language Models

2402

Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models

The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits

When Scaling Meets LLM Finetuning: The Effect of Data, Model and Finetuning Method

MobiLlama: Towards Accurate and Lightweight Fully Transparent GPT

Training Neural Networks from Scratch with Parallel Low-Rank Adapters

FuseChat: Knowledge Fusion of Chat Models

MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases

Efficient and Effective Vocabulary Expansion Towards Multilingual Large Language Models

KorNAT: LLM Alignment Benchmark for Korean Social Values and Common Knowledge

Chain of Thought Empowers Transformers to Solve Inherently Serial Problems

DoRA: Weight-Decomposed Low-Rank Adaptation

Accurate LoRA-Finetuning Quantization of LLMs via Information Retention

More Agents Is All You Need

ScreenAI: A Vision-Language Model for UI and Infographics Understanding

DISTILLM: Towards Streamlined Distillation for Large Language Models

ReLU$^2$ Wins: Discovering Efficient Activation Functions for Sparse LLMs

2401

DressCode: Autoregressively Sewing and Generating Garments from Text Guidance

Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling

SliceGPT: Compress Large Language Models by Deleting Rows and Columns

Demystifying Chains, Trees, and Graphs of Thoughts

Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

Tuning Language Models by Proxy

Heterogeneous LoRA for Federated Fine-tuning of On-Device Foundation Models

RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models

2312

SOLAR 10.7B: Scaling Large Language Models with Simple yet Effective Depth Up-Scaling

PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU

LLM in a flash: Efficient Large Language Model Inference with Limited Memory

LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin

Mamba: Linear-Time Sequence Modeling with Selective State Spaces

2311

Diffusion Models Without Attention

Knowledge Transfer from Vision Foundation Models for Efficient Training of Small Task-specific Models

Probabilistic Speech-Driven 3D Facial Motion Synthesis: New Benchmarks, Methods, and Applications

Swallowing the Bitter Pill: Simplified Scalable Conformer Generation

MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training

RELIC: Investigating Large Language Model Responses using Self-Consistency

Direct2.5: Diverse 3D Content Creation via Multi-view 2.5D Diffusion

LQ-LoRA: Low-rank Plus Quantized Matrix Decomposition for Efficient Language Model Finetuning

MultiLoRA: Democratizing LoRA for Better Multi-Task Learning

Tied-Lora: Enhancing parameter efficiency of LoRA with weight tying

S-LoRA: Serving Thousands of Concurrent LoRA Adapters

FlashDecoding++: Faster Large Langauge Model Inference on GPUs

Efficient LLM Inference on CPUs

2310

EELBERT: Tiny Models through Dynamic Embeddings

Accelerating LLaMA Inference by Enabling Intermediate Layer Decoding via Instruction Tuning with LITE

LoRAShear: Efficient Large Language Model Structured Pruning and Knowledge Recovery

FP8-LM: Training FP8 Large Language Models

Large Language Models as Generalizable Policies for Embodied Tasks

LLM-FP4: 4-Bit Floating-Point Quantized Transformers

Matryoshka Diffusion Models

We are Who We Cite: Bridges of Influence Between Natural Language Processing and Other Academic Fields

CLIP meets Model Zoo Experts: Pseudo-Supervision for Visual Enhancement

BitNet: Scaling Transformers for Large Language Models

When Can Transformers Reason With Abstract Symbols?

LoftQ: LoRA-Fine-Tuning-Aware Quantization for Large Language Models

Pseudo-Generalized Dynamic View Synthesis from a Video

QLLM: Accurate and Efficient Low-Bitwidth Quantization for Large Language Models

Generative Modeling with Phase Stochastic Bridges

Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning

iTransformer: Inverted Transformers Are Effective for Time Series Forecasting

JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modeling

Chat Vector: A Simple Approach to Equip LLMs with Instruction Following and Model Alignment in New Languages

ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models

Improved Baselines with Visual Instruction Tuning

Compressing LLMs: The Truth is Rarely Pure and Never Simple

2309

Efficient Streaming Language Models with Attention Sinks

Guiding Instruction-based Image Editing via Multimodal Large Language Models

Vision Transformers Need Registers

QA-LoRA: Quantization-Aware Low-Rank Adaptation of Large Language Models

Efficient Memory Management for Large Language Model Serving with PagedAttention

Pushing Mixture of Experts to the Limit: Extremely Parameter Efficient MoE for Instruction Tuning

2308

Reducing shared memory footprint to leverage high throughput on Tensor Cores and its flexible API extension library

Fast Feedforward Networks

EdgeMoE: Fast On-Device Inference of MoE-based Large Language Models

OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models

2307

ZeroQuant-FP: A Leap Forward in LLMs Post-Training W4A8 Quantization Using Floating-Point Formats

FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning

2306

MOFI: Learning Image Representation from Noisy Entity Annotated Images

SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression

TIES-Merging: Resolving Interference When Merging Models

AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

2305

Direct Preference Optimization: Your Language Model is Secretly a Reward Model

Manifold Diffusion Fields

QLoRA: Efficient Finetuning of Quantized LLMs

RWKV: Reinventing RNNs for the Transformer Era

Tree of Thoughts: Deliberate Problem Solving with Large Language Models

SpecInfer: Accelerating Generative Large Language Model Serving with Tree-based Speculative Inference and Verification

MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers

Fast Distributed Inference Serving for Large Language Models

2304

Visual Instruction Tuning

2303

Sigmoid Loss for Language Image Pre-Training

ZeroQuant-V2: Exploring Post-training Quantization in LLMs from Comprehensive Study to Low Rank Compensation

FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU

2302

Full Stack Optimization of Transformer Inference: a Survey

2301

Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture

2211

Fast Inference from Transformers via Speculative Decoding

SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models

Who Says Elephants Can't Run: Bringing Large Scale MoE Models into Cloud Scale Production

2210

ByteTransformer: A High-Performance Transformer Boosted for Variable-Length Inputs

2209

FP8 Formats for Deep Learning

2208

LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale

2206

ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers

2205

FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning

Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning

2204

Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

2203

A Survey of Multi-Tenant Deep Learning Inference on GPU

2202

cosFormer: Rethinking Softmax in Attention

2106

LibShalom: Optimizing Small and Irregular-Shaped Matrix Multiplications on ARMv8 Multi-Cores

LoRA: Low-Rank Adaptation of Large Language Models

2104

RoFormer: Enhanced Transformer with Rotary Position Embedding

2010

TurboTransformers: An Efficient GPU Serving System For Transformer Models

2009

Flexible Performant GEMM Kernels on GPUs

2005

BiQGEMM: Matrix Multiplication with Lookup Table For Binary-Coding-based Quantized DNNs

1805

Online normalizer calculation for softmax

1803

NVIDIA Tensor Core Programmability, Performance & Precision

1706

Attention Is All You Need