← Back to all news
arXiv Computation and Language By Gaurab Baral, Mimansak Nepal, Grishma Basnet

NEPSEE: A Double-Annotated Benchmark for Nepali Sentiment Analysis

Read the original on arXiv Computation and Language →

NEPSEE is a Nepali sentiment analysis benchmark created by having two annotators independently label 5,614 tweets from Nepali public figures on a five‑point scale. Only 3,189 sentences with identical polarity are kept for a three‑class version, while 766 sentences with adjacent labels are used in a five‑point task. Baseline results show NepBERTa achieving a macro‑F1 of 0.877 on the three‑class benchmark, with commercial LLMs performing similarly; on the five‑point task, Claude Opus 5.5 outperforms encoders by 0.10–0.12 macro‑F1, and a tokenizer audit reveals that restoring Devanagari vowel signs can significantly alter performance.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

Related stories

arXiv Computation and Language
Sep 25

Two Emojis of Difference: What Multilingual Affective Generation Benchmarks Actually Measure

The paper audits a multilingual affective generation benchmark that uses emoji summaries for Bangla, English, and Hindi sentences. It finds that the benchmark’s conclusions are largely artifacts of the measurement instrument, with no system significantly outperforming another when annotators are treated as random factors. The study shows that annotator identity and output length drive most variance, and proposes a new stable metric called emoji‑affect decodability.

By Fardeen Sadab, Adib Sakhawat
llmsbenchmarks
More like this →
arXiv Machine Learning
Aug 28

Vowel Signs Are Not Letters: A Pre-tokenization Ceiling on Multilingual Tokenizer Fertility

The paper shows that HuggingFace’s ByteLevel pre‑tokenizer, which treats a word as a sequence of Unicode letters, splits abugida scripts at every vowel sign, creating a training‑free lower bound on tokenizer fertility. Across 26 languages, all 17 abugidas exhibit increased token counts (up to 9×), while Latin, Cyrillic, Hangul, and Han remain unchanged. The authors demonstrate that correcting the character class reduces Nepali token counts, improves model performance, and that this issue is widespread in popular HuggingFace models.

By Sajal Regmi, Siddhartha Pudasaini, Chetan Phakami Pun
llmsnlp
More like this →
arXiv AI
Aug 20

NE-BERT: A Multilingual Language Model for Nine Northeast Indian Languages

NE‑BERT is a multilingual encoder trained on about 8.3 million sentences from nine Northeast Indian languages plus Hindi and English. Using weighted sampling and a custom SentencePiece tokenizer, it achieves significantly lower perplexity than IndicBERT‑V2, MuRIL, and mBERT, and improves tokenization fertility. The model also addresses vocabulary fragmentation in extremely low‑resource languages through aggressive upsampling, and its effectiveness is validated on part‑of‑speech tagging for three of the languages.

By Badal Nyalang
llmsnlp
More like this →