Fine-Tuning Transformers vs. Few-Shot LLMs for Bangla NLP
25 Jun 2026
Compare Bangla PLM fine-tuning via AdamW with zero-shot and few-shot (5 to 15-shot) generative prompt engineering.
Inside the Motamot Dataset: Annotation & Quality Control for Bangla NLP
24 Jun 2026
Explore the curation, structure, and quality control of the Motamot dataset—a manually annotated 7,058-instance corpus for Bengali political opinion mining.
Bangla NLP Architecture Guide: Pre-trained Transformers vs. Frontier LLMs
24 Jun 2026
iscover how localized PLMs like BanglaBERT and SahajBERT match up against the multi-lingual context windows of Gemini 1.5 Pro and GPT-3.5 Turbo.
Hybrid NLP & LLM Sentiment Analysis: Multi-Domain Literature Review
24 Jun 2026
Compare traditional ML, hybrid CNN-LSTMs, and LLM Chain-of-Thought weak labeling techniques.
Motamot Dataset: Benchmarking LLMs vs PLMs in Bangla Political NLP
24 Jun 2026
Explore how few-shot learning eliminates hallucinations and drives Gemini 1.5 Pro to outperform fine-tuned PLMs.
LLMs vs Transformers: Bengali Political Sentiment Analysis Benchmark
23 Jun 2026
Learn how few-shot learning drives Gemini 1.5 Pro to a 96.33% accuracy rate.
References for Web-Scale Information Retrieval Challenges
2 Jul 2025
A list of scholarly references at the intersection of deep learning in information retrieval, large-scale approximate nearest neighbor search
Navigating Skew: Addressing Language & Domain Biases in Web Data
2 Jul 2025
Explore the challenges posed by high-skewed language and topic distributions in web data, acknowledging potential model biases
Mind the Gap: End-to-End Quality Drop with ANN in Web Search AI
2 Jul 2025
Discover how integrating ANN indices leads to a substantial drop in final retrieval quality compared to brute-force search