BestHub
Sign in
SmartApe AI
Author

SmartApe AI

SmartApe's AI technology content account, focusing on cutting-edge AI developments and open-source frontlines.

70
Articles
1
Likes
97
Views
0
Comments
Recent Articles

Latest from SmartApe AI

70 recent articles
SmartApe AI
SmartApe AI
Oct 29, 2026 · Industry Insights

From Korean Major to Google Gemini: A Linguist's Path to Top AI Research

Shanghai International Studies University Korean major graduate Fangru Lin joined Google DeepMind's Gemini pre-training team as a research scientist after publishing first-author papers at ICML and ACL on LLM reasoning, planning, and dialect fairness, following a linguistics PhD at Oxford and internships at Microsoft and Google.

AI TalentCross-Representation GeneralizationDialect Fairness
0 likes · 7 min read
From Korean Major to Google Gemini: A Linguist's Path to Top AI Research
SmartApe AI
SmartApe AI
Oct 29, 2026 · Artificial Intelligence

CMU's Hermit: Unified Framework for Fair Harness Evolution Comparison

CMU researchers open-source Hermit, a unified training framework that enables fair comparison of recursive self-improvement methods by externalizing learning via pluggable Harness Adapters, supporting algorithms like GEPA and SkillOpt, and providing a tutorial for continuous agent improvement in tools like Codex and Claude Code.

Agent Continuous LearningCMU ResearchCodex Integration
0 likes · 8 min read
CMU's Hermit: Unified Framework for Fair Harness Evolution Comparison
SmartApe AI
SmartApe AI
Oct 28, 2026 · Artificial Intelligence

VLDA: Learning Compact State Deltas Beats Full Future Prediction for Robot Manipulation

Harbin Institute of Technology and Baidu propose VLDA, a robotic manipulation paradigm that learns compact task-relevant state changes (Delta) instead of full future prediction, achieving 98.5% on LIBERO, 84.5% on LIBERO-Plus, and 73% real-robot success rates by using a Delta variable as an explicit transition interface between vision-language understanding and action generation.

LoongForgeVLDAembodied AI
0 likes · 21 min read
VLDA: Learning Compact State Deltas Beats Full Future Prediction for Robot Manipulation
SmartApe AI
SmartApe AI
Oct 28, 2026 · Artificial Intelligence

Qwen 4 Imminent: Leaked Code & Anonymous Models Reveal Architecture Details

Community clues suggest Alibaba's Qwen 4 is near release, with inference optimizations in vLLM and llama.cpp, architectural details like a 27B dense model with 5120 hidden dim, 24 query heads, 4 KV heads, 4 residual streams, MTP, and QSA sparse attention, while anonymous Arena models Kalmia and Kerria are speculated as Qwen 4 variants showing strong code and 3D generation.

AlibabaArena benchmarkFP8 quantization
0 likes · 7 min read
Qwen 4 Imminent: Leaked Code & Anonymous Models Reveal Architecture Details
SmartApe AI
SmartApe AI
Oct 28, 2026 · Artificial Intelligence

UW PhD Joins NUS, CoRL 2026 Spotlight: $6K Robot Achieves Embodied Reasoning

University of Washington PhD Jianfei Duan joins NUS as Presidential Young Professor, founding MAGIC Lab; his CoRL 2026 Spotlight paper MolmoAct2 and Oral paper MolmoB0T demonstrate low-cost robots ($6K) performing embodied reasoning, building on prior work in robot learning, failure diagnosis, and generalization benchmarks.

CoRL 2026Failure DiagnosisGeneralization Benchmarks
0 likes · 6 min read
UW PhD Joins NUS, CoRL 2026 Spotlight: $6K Robot Achieves Embodied Reasoning
SmartApe AI
SmartApe AI
Oct 28, 2026 · Artificial Intelligence

DeepSeek Flash Fingerprint Changes Again, V5 Names Surface in Test Repo

On Oct 27, DeepSeek's Flash model fingerprints changed for the second time in 10 days, with three model aliases switching to a new system fingerprint. Community testing shows stable new fingerprint but no behavioral differences yet. Meanwhile, DeepSeek Harness test files contain mock V5 model names, fueling speculation about upcoming releases, though V4 Pro remains unchanged and official changelog hasn't acknowledged the shift.

AI model deploymentAPI monitoringDeepSeek
0 likes · 8 min read
DeepSeek Flash Fingerprint Changes Again, V5 Names Surface in Test Repo
SmartApe AI
SmartApe AI
Oct 25, 2026 · Artificial Intelligence

Wafer-Scale AI Chips: Co-Designing for 4x Token-per-Dollar Gains

Tsinghua's Hu Yang presents wafer-scale chip co-design to overcome communication, power, and density walls, achieving 1.21x token-per-dollar over NVL72-GB200 with domestic process and 4x with 4nm, via integrated architecture, interconnect, fault tolerance, and thermal management.

AI hardwarechiplet integrationdomestic process
0 likes · 16 min read
Wafer-Scale AI Chips: Co-Designing for 4x Token-per-Dollar Gains
SmartApe AI
SmartApe AI
Oct 24, 2026 · Artificial Intelligence

ByteDance TraceDance: 250K Real Trajectories Expose Frontier Model Behavioral Failures

ByteDance's TraceDance system automatically converts 252,557 real-world agent deployment trajectories into 107 behavioral benchmarks with 4,125 test instances, evaluating nine frontier models including Claude, GPT, and DeepSeek, which achieve only a 26.7% average pass rate on decision-point continuation tasks, revealing gaps between task completion and critical behavioral decisions.

Agent EvaluationBehavioral BenchmarksByteDance
0 likes · 17 min read
ByteDance TraceDance: 250K Real Trajectories Expose Frontier Model Behavioral Failures