analiza aiqbench
Emo-Jev presents a training-free probabilistic framework that decomposes emotion classification tasks into atomic judgments, offering a cost-effective and low-latency alternative to standard large language models. By aggregating probabilities from multiple reasoning paths, this approach demonstrates competitive performance across sentiment analysis and sarcasm detection without requiring fine-tuning. This work highlights the potential of structured probabilistic interfaces as viable substitutes for heavy LLM deployments in specific NLP classification domains.
Emo-Jev: Probabilistic Reasoning for Emotion Classification with Jev
arXiv:2610.08829v1 Announce Type: new Abstract: Jev offers an alternative interface for language understanding: given an input and predefined questions, it returns probabilistic decisions rather than free-form responses. Whether this interface can support effective reasoning for text classification against leading LLMs remains an open questions. We introduce Emo-Jev, a training-free framework with two complementary implementations. Emo-Jev-D decomposes classification into task-specific atomic judgments and composes their probabilities into a final prediction. Emo-Jev-SC constructs multiple judgment paths from complementary perspectives and aggregates their predictions into a consensus decision. We evaluate Emo-Jev on eight datasets spanning sentiment analysis, emotion recognition, sarcasm detection and humor detection, comparing against direct Jev classification and five SoTA LLMs under input/output and chain-of-thought reasoning. Standard Jev achieves 62.93\% average macro-F1 versus 67.28\% for the strongest LLM baseline, with lower observed latency and generally lower cost.
Abstract:Jev offers an alternative interface for language understanding: given an input and predefined questions, it returns probabilistic decisions rather than free-form responses. Whether this interface can support effective reasoning for text classification against leading LLMs remains an open questions. We introduce Emo-Jev, a training-free framework with two complementary implementations. Emo-Jev-D decomposes classification into task-specific atomic judgments and composes their probabilities into a final prediction. Emo-Jev-SC constructs multiple judgment paths from complementary perspectives and aggregates their predictions into a consensus decision. We evaluate Emo-Jev on eight datasets spanning sentiment analysis, emotion recognition, sarcasm detection and humor detection, comparing against direct Jev classification and five SoTA LLMs under input/output and chain-of-thought reasoning. Standard Jev achieves 62.93\% average macro-F1 versus 67.28\% for the strongest LLM baseline, with lower observed latency and generally lower cost.
Computation and Language (cs.CL)
arXiv:2610.08829 [cs.CL]
(or arXiv:2610.08829v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2610.08829 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 224 cuvinte
Subiecte conexe
The user in the LLM might be a moral patient too (a strange implication of the PSM and simulators)
LessWrong · 15 oct. 2026
CLARK: Closed-loop Learning for Adaptive Reasoning over Knowledge Graphs
arXiv · 23 iul. 2026
A Neuro-Symbolic Approach for Probabilistic Reasoning on Graph Data
arXiv · 20 iul. 2026
Speech-based Psychological Crisis Assessment using LLMs
arXiv · 1 oct. 2026