Amazon

company

Verified

https://aws.amazon.com/

AI & ML interests

Scalable Artificial Intelligence

Recent Activity

zancato published a model about 17 hours ago

amazon/Mamba2-primed-HQwen3-8B-Instruct

zancato published a model about 17 hours ago

amazon/GKA-primed-HQwen3-8B-Instruct

zancato published a model about 17 hours ago

amazon/GDN-primed-HQwen3-8B-Instruct

View all activity

Papers

Scalable Prompt Routing via Fine-Grained Latent Task Discovery

Perceptio: Perception Enhanced Vision Language Models via Spatial Token Generation

View all Papers

amazon 's Papers 22

Submitted by

Yunyi Zhang

Scalable Prompt Routing via Fine-Grained Latent Task Discovery

amazon

2

Submitted by

Rui Zhao

Perceptio: Perception Enhanced Vision Language Models via Spatial Token Generation

amazon

Submitted by

Jiarui Yao

Supervised Fine-Tuning versus Reinforcement Learning: A Study of Post-Training Methods for Large Language Models

amazon

Submitted by

Chunyuan Deng

ByteFlow: Language Modeling through Adaptive Byte Compression without a Tokenizer

amazon

2

Submitted by

taesiri

DDiT: Dynamic Patch Scheduling for Efficient Diffusion Transformers

amazon

Submitted by

CL Yu

Approximation of Log-Partition Function in Policy Mirror Descent Induces Implicit Regularization for LLM Post-Training

amazon

Submitted by

Awsaf

LikeBench: Evaluating Subjective Likability in LLMs for Personalization

amazon

2

Submitted by

taesiri

WebCoach: Self-Evolving Web Agents with Cross-Session Memory Guidance

amazon

Submitted by

Yifan Jiang

VIDEOP2R: Video Understanding from Perception to Reasoning

amazon

5

Submitted by

Soyeong

Adaptive Multi-Agent Response Refinement in Conversational Systems

amazon

Submitted by

Song

Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs

amazon

2

Submitted by

Abdul Fatir Ansari

Chronos-2: From Univariate to Universal Forecasting

amazon

Submitted by

taicheng guo

MTSQL-R1: Towards Long-Horizon Multi-Turn Text-to-SQL via Agentic Training

amazon

Submitted by

Weijie Xu

The Personalization Trap: How User Memory Alters Emotional Reasoning in LLMs

amazon

Submitted by

Zhenhailong Wang

Multimodal Policy Internalization for Conversational Agents

amazon

Submitted by

Soyeong

When Thoughts Meet Facts: Reusable Reasoning for Long-Context LMs

amazon

2

Submitted by

Jiaru Zou

TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning

amazon

3

From Frames to Clips: Efficient Key Clip Selection for Long-Form Video Understanding

amazon

Training Large Language Models To Reason In Parallel With Global Forking Tokens

amazon

Submitted by

Gabriele Berton

CompLLM: Compression for Long Context Q&A

amazon

4

FLAMES: Improving LLM Math Reasoning via a Fine-Grained Analysis of the Data Synthesis Pipeline

amazon

1

Submitted by

Weijie Xu

Quantifying Fairness in LLMs Beyond Tokens: A Semantic and Statistical Perspective

amazon