🌐 Entity

Hallucination

Perception that only seems real

📊 Rating

3 news mentions · 👍 0 likes · 👎 0 dislikes

📌 Topics

Artificial Intelligence (3)
Computer Vision (2)
Machine Learning (1)
Linguistic Analysis (1)
Model Performance (1)
Model Evaluation (1)
AI Safety (1)

🏷️ Keywords

Vision-Language Models (2) · Hallucination (2) · Computer Vision (2) · Spatial Credit Redistribution (1) · Transformer (1) · AI Research (1) · Large Language Models (1) · Hallucinations (1) · Query Features (1) · Linguistic Complexity (1) · Computational Linguistics (1) · Model Accuracy (1) · Query Optimization (1) · VAUQ (1) · Uncertainty Quantification (1) · Model Evaluation (1) · Self-Evaluation (1) · AI Safety (1)

📖 Key Information

A hallucination is a perception in the absence of an external context stimulus that has the compelling sense of reality. They are distinguishable from several related phenomena, such as dreaming (REM sleep), which does not involve wakefulness; pseudohallucination, which does not mimic real perception, and is accurately perceived as unreal; illusion, which involves distorted or misinterpreted real perception; and mental imagery, which does not mimic real perception, and is under voluntary control. Hallucinations also differ from "delusional perceptions", in which a correctly sensed and interpreted stimulus (i.e., a real perception) is given some additional significance.

📰 Related News (3)

🇺🇸 Beyond Dominant Patches: Spatial Credit Redistribution For Grounded Vision-Language Models (2026-02-27)
arXiv:2602.22469v1 Announce Type: cross Abstract: Vision-language models (VLMs) frequently hallucinate objects absent from the input image. We trace ...
🇺🇸 What Makes a Good Query? Measuring the Impact of Human-Confusing Linguistic Features on LLM Performance (2026-02-25)
arXiv:2602.20300v1 Announce Type: cross Abstract: Large Language Model (LLM) hallucinations are usually treated as defects of the model or its decodi...
🇺🇸 VAUQ: Vision-Aware Uncertainty Quantification for LVLM Self-Evaluation (2026-02-25)
arXiv:2602.21054v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) frequently hallucinate, limiting their safe deployment in real...

🔗 Entity Intersection Graph

People and organizations frequently mentioned alongside Hallucination:

🌐
Computer vision · 2 shared articles
🌐
Uncertainty quantification · 1 shared articles
🌐
Computational linguistics · 1 shared articles
🌐
Large language model · 1 shared articles
Transformer · 1 shared articles