Lost in Backpropagation: The LM Head is a Gradient Bottleneck
Paper
• 2603.10145 • Published
• 4
Cornell University
university
Verified
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Lost in Backpropagation: The LM Head is a Gradient Bottleneck
YOLOE-26: Integrating YOLO26 with YOLOE for Real-Time Open-Vocabulary Instance Segmentation
nthngdy
submitted a paper to Daily Papers 1 day ago
RanjanSapkota
authored a paper about 1 month ago
RanjanSapkota
submitted a paper to Daily Papers about 1 month ago
RanjanSapkota
authored 2 papers 3 months ago
The SAM2-to-SAM3 Gap in the Segment Anything Model Family: Why Prompt-Based Expertise Fails in Concept-Driven Image Segmentation
Paper
• 2512.06032 • Published
• 1
Generalization vs. Specialization: Evaluating Segment Anything Model (SAM3) Zero-Shot Segmentation Against Fine-Tuned YOLO Detectors
Paper
• 2512.11884 • Published
• 2
osayamenja
authored a paper 4 months ago
RanjanSapkota
authored 3 papers 5 months ago
YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection
Paper
• 2509.25164 • Published
Ultralytics YOLO Evolution: An Overview of YOLO26, YOLO11, YOLOv8 and YOLOv5 Object Detectors for Computer Vision and Pattern Recognition
Paper
• 2510.09653 • Published
Object Detection with Multimodal Large Vision-Language Models: An In-depth Review
Paper
• 2508.19294 • Published
• 1
RanjanSapkota
authored 3 papers 8 months ago
3D Reconstruction and Information Fusion between Dormant and Canopy Seasons in Commercial Orchards Using Deep Learning and Fast GICP
Paper
• 2507.01912 • Published
Orchestrator-Agent Trust: A Modular Agentic AI Visual Classification System with Trust-Aware Orchestration and RAG-Based Reasoning
Paper
• 2507.10571 • Published
• 3
Thinking Beyond Tokens: From Brain-Inspired Intelligence to Cognitive Foundations for Artificial General Intelligence and its Societal Impact
Paper
• 2507.00951 • Published
• 24
RanjanSapkota
authored 2 papers 9 months ago
DaiYijia
authored 5 papers 9 months ago
End-to-end Training for Recommendation with Language-based User Profiles
Paper
• 2410.18870 • Published
Language-Based User Profiles for Recommendation
Paper
• 2402.15623 • Published
Sample Efficient Preference Alignment in LLMs via Active Exploration
Paper
• 2312.00267 • Published
Representation Learning in Low-rank Slate-based Recommender Systems
Paper
• 2309.08622 • Published
Pre-trained Large Language Models Learn Hidden Markov Models In-context
Paper
• 2506.07298 • Published
• 26
RanjanSapkota
authored a paper 10 months ago