Mitigating hallucinations in large vision-language models (LVLMs) remains an open problem. Recent benchmarks do not address hallucinations in open-ended free-form responses, which we term "Type I hallucinations". Instead, they focus on hallucinations responding to very specific question formats -- typically a multiple-choice response regarding a particular object or attribute -- which we term "Type II hallucinations". Additionally, such benchmarks often require external API calls to models which are subject to change. In practice, we observe that a reduction in Type II hallucinations does not lead to a reduction in Type I hallucinations but rather that the two forms of hallucinations are often anti-correlated. To address this, we propose THRONE, a novel object-based automatic framework for quantitatively evaluating Type I hallucinations in LVLM free-form outputs. We use public language models (LMs) to identify hallucinations in LVLM responses and compute informative metrics. By evaluating a large selection of recent LVLMs using public datasets, we show that an improvement in existing metrics do not lead to a reduction in Type I hallucinations, and that established benchmarks for me
In this paper, we present a MLP-like architecture for sequential recommendation, namely TriMLP, with a novel Triangular Mixer for cross-token communications. In designing Triangular Mixer, we simplify the cross-token operation in MLP as the basic matrix multiplication, and drop the lower-triangle neurons of the weight matrix to block the anti-chronological order connections from future tokens. Accordingly, the information leakage issue can be remedied and the prediction capability of MLP can be fully excavated under the standard auto-regressive mode. Take a step further, the mixer serially alternates two delicate MLPs with triangular shape, tagged as global and local mixing, to separately capture the long range dependencies and local patterns on fine-grained level, i.e., long and short-term preferences. Empirical study on 12 datasets of different scales (50K\textasciitilde 10M user-item interactions) from 4 benchmarks (Amazon, MovieLens, Tenrec and LBSN) show that TriMLP consistently attains promising accuracy/efficiency trade-off, where the average performance boost against several state-of-the-art baselines achieves up to 14.88% with 8.65% less inference cost.
Early reports indicate there may be another case, but spread is thought to be localized
A distant galaxy nicknamed Shadow Blaster may have revealed a surprising source of cosmic neutrinos: extreme star formation instead of a supermassive black hole。 The discovery suggests that hidden, dust-filled starburst galaxies could account for a significant fraction of the Universe’s high-energy neutrinos
It only works for a few divisions thanks to a lot of added materials
Insiders say Sam Altman is in active talks with the Trump administration
New York City’s Summer of Ludd festival is teaching people how to live offline
Scientists are raising concerns that we may be overlooking evidence of extraterrestrial life even when it is present。 Hidden biosignatures, limitations in detection technology, and assumptions about what life should look like can all create dangerous false negatives。 The researchers say future missions should focus not only on finding life, but als
A surprisingly simple fuel modification could help tackle one of diesel engines’ biggest problems: pollution。 Researchers reviewing studies from around the world found that mixing small amounts of water into diesel fuel can dramatically reduce harmful emissions, including nitrogen oxides and soot, while maintaining or even improving engine efficien
Scientists have uncovered a surprising connection between quantum gravity and an exotic quantum state of matter that could explain why the universe isn’t expanding wildly fast。 The study suggests that the very shape of space-time may protect the cosmological constant from disruptive quantum effects
Walkthrough experience includes visits to stars, exoplanets, and observatories
A rare meteorite has revealed evidence of a massive lost world that once orbited the young Sun before being destroyed in a catastrophic collision。 The discovery suggests some early planets formed from dramatically different materials than Earth and Mars, rewriting part of the solar system’s origin story
A clever nanoscale redesign may have solved one of superconductivity’s biggest problems。 Researchers in Sweden discovered that by subtly sculpting the surface beneath an ultrathin superconducting material, they could make it stay superconducting at higher temperatures and under much stronger magnetic fields
NASA’s upgraded Cold Atom Lab is turning the International Space Station into a frontier for quantum research, creating ultra-cold matter that behaves in astonishing ways。 The experiments could unlock new discoveries about the universe while paving the way for powerful future technologies in space and on Earth
Researchers developed a Wordle-solving strategy that succeeds 99% of the time by focusing on information gain rather than likely answers。 The method uses Shannon entropy to identify guesses that reveal the most about the hidden word。 Each guess is designed to slash uncertainty and narrow the possibilities faster
NASA awarded Rocket Lab deals for three dedicated launches using the company's Electron rocket
Doctors find grey fluid and dead, metallic flesh inside poisoned woman's hip
Money laundering has become one of the most relevant criminal activities in modern societies, as it causes massive financial losses for governments, banks and other institutions. Detecting such activities is among the top priorities when it comes to financial analysis, but current approaches are often costly and labor intensive partly due to the sheer amount of data to be analyzed. Hence, there is a growing need for automatic anti-money laundering systems to assist experts. In this work, we propose DELATOR, a novel framework for detecting money laundering activities based on graph neural networks that learn from large-scale temporal graphs. DELATOR provides an effective and efficient method for learning from heavily imbalanced graph data, by adapting concepts from the GraphSMOTE framework and incorporating elements of multi-task learning to obtain rich node embeddings for node classification. DELATOR outperforms all considered baselines, including an off-the-shelf solution from Amazon AWS by 23% with respect to AUC-ROC. We also conducted real experiments that led to the discovery of 7 new suspicious cases among the 50 analyzed ones, which have been reported to the authorities.
Astronomers studying the rare supernova SN 2021yfj discovered material from one of the deepest layers of a dying star, providing a rare look at its hidden interior。 The finding confirms key theories about how massive stars forge the elements that help build planets, worlds, and life