NLP

Arctic-ABSA: Smarter, multilingual aspect sentiment with reasoning

Kari Jaaskelainen

08 Jan 2026 — 1 min read

Ever wish reviews told you exactly how people feel about the camera, battery, or price? That's aspect-based sentiment analysis (ABSA).

Arctic-ABSA is a new family of reasoning-infused models built for real-world ABSA. Trained on a large blend of public and carefully generated synthetic data—about 20× the classic SemEval14 size—they:

Recognize five sentiments: positive, negative, neutral, mixed, and unknown.
Judge both per-aspect opinions and overall text sentiment.
Work across multiple languages with one model.

By injecting step-by-step reasoning (including a new pretraining method for encoder models), Arctic-ABSA generalizes better and stays accurate when the data shifts.

Results: a 395M-parameter encoder and an 8B-parameter decoder set new state-of-the-art ABSA scores, beating leading general LLMs (e.g., GPT-4o, Claude 3.5 Sonnet) by up to 10 percentage points. One multilingual model maintains 87–91% accuracy across six languages without hurting English.

Plus, the team releases ABSA-mix, a large benchmark spanning 17 public datasets and 92 domains.

Paper: https://arxiv.org/abs/2601.03940v1

Paper: https://arxiv.org/abs/2601.03940v1

Register: https://www.AiFeta.com

NLP SentimentAnalysis ABSA AI LLM Multilingual Reasoning Benchmark Research

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added

Read more

Meet GRACE: a moral governor for safer, more transparent AI

Why some fine-tuned LLMs miss phishing—and how to fix it

AI that explains itself—by following the science

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A