Close Menu
Ztoog
    What's Hot
    The Future

    Russell T. Davies Explains Doctor Who’s Big Davros Retcon

    Crypto

    Crypto VC exits were low in Q4 2023, Phantom MAU’s reach new highs and spot bitcoin ETF volumes are still rising

    The Future

    I’m trying so hard not to gush over Teenage Engineering’s latest gadget

    Important Pages:
    • About Us
    • Contact us
    • Privacy Policy
    • Terms & Conditions
    Facebook X (Twitter) Instagram Pinterest
    Facebook X (Twitter) Instagram Pinterest
    Ztoog
    • Home
    • The Future

      Tencent hires WizardLM team, a Microsoft AI group with an odd history

      Today’s NYT Connections Hints, Answers for May 12, #701

      OPPO launches A5 Pro 5G: Premium features at a budget price

      How I Turn Unstructured PDFs into Revenue-Ready Spreadsheets

      Is it the best tool for 2025?

    • Technology

      Apple will use AI and user data in iOS 19 to extend iPhone battery life

      Today’s NYT Wordle Hints, Answer and Help for May 12, #1423

      What It Is and Why It Matters—Part 1 – O’Reilly

      Ensure Hard Work Is Recognized With These 3 Steps

      Cicada map 2025: Where will Brood XIV cicadas emerge this spring?

    • Gadgets

      “Google wanted that”: Nextcloud decries Android permissions as “gatekeeping”

      Google Tests Automatic Password-to-Passkey Conversion On Android

      Maono Caster G1 Neo & PD200X Review: Budget Streaming Gear for Aspiring Creators

      Apple plans to split iPhone 18 launch into two phases in 2026

      Upgrade your desk to Starfleet status with this $95 USB-C hub

    • Mobile

      Fitbit’s AI experiments just leveled up with 3 new health tracking features

      Motorola’s Moto Watch needs to start living up to the brand name

      Samsung Galaxy S25 Edge promo materials leak

      What are people doing with those free T-Mobile lines? Way more than you’d expect

      Samsung doesn’t want budget Galaxy phones to use exclusive AI features

    • Science

      Ancient three-eyed ‘sea moth’ used its butt to breathe

      Intelligence on Earth Evolved Independently at Least Twice

      Nothing is stronger than quantum connections – and now we know why

      Failed Soviet probe will soon crash to Earth – and we don’t know where

      Trump administration cuts off all future federal funding to Harvard

    • AI

      Study shows vision-language models can’t handle queries with negation words | Ztoog

      How a new type of AI is helping police skirt facial recognition bans

      Hybrid AI model crafts smooth, high-quality videos in seconds | Ztoog

      How to build a better AI benchmark

      Q&A: A roadmap for revolutionizing health care through data-driven innovation | Ztoog

    • Crypto

      HashKey Group Announces Launch of HashKey Global MENA with VASP License in UAE

      Ethereum Breaks Key Resistance In One Massive Move – Higher High Confirms Momentum

      ‘The Big Short’ Coming For Bitcoin? Why BTC Will Clear $110,000

      Bitcoin Holds Above $95K Despite Weak Blockchain Activity — Analytics Firm Explains Why

      eToro eyes US IPO launch as early as next week amid easing concerns over Trump’s tariffs

    Ztoog
    Home » Study shows vision-language models can’t handle queries with negation words | Ztoog
    AI

    Study shows vision-language models can’t handle queries with negation words | Ztoog

    Facebook Twitter Pinterest WhatsApp
    Study shows vision-language models can’t handle queries with negation words | Ztoog
    Share
    Facebook Twitter LinkedIn Pinterest WhatsApp

    Imagine a radiologist analyzing a chest X-ray from a brand new affected person. She notices the affected person has swelling within the tissue however doesn’t have an enlarged coronary heart. Looking to hurry up prognosis, she may use a vision-language machine-learning mannequin to seek for studies from comparable sufferers.

    But if the mannequin mistakenly identifies studies with each circumstances, the most certainly prognosis may very well be fairly completely different: If a affected person has tissue swelling and an enlarged coronary heart, the situation could be very prone to be cardiac associated, however with no enlarged coronary heart there may very well be a number of underlying causes.

    In a brand new examine, MIT researchers have discovered that vision-language models are extraordinarily prone to make such a mistake in real-world conditions as a result of they don’t perceive negation — words like “no” and “doesn’t” that specify what is fake or absent. 

    “Those negation words can have a very significant impact, and if we are just using these models blindly, we may run into catastrophic consequences,” says Kumail Alhamoud, an MIT graduate scholar and lead writer of this examine.

    The researchers examined the flexibility of vision-language models to determine negation in picture captions. The models typically carried out in addition to a random guess. Building on these findings, the crew created a dataset of photographs with corresponding captions that embody negation words describing lacking objects.

    They present that retraining a vision-language mannequin with this dataset results in efficiency enhancements when a mannequin is requested to retrieve photographs that don’t include sure objects. It additionally boosts accuracy on a number of selection query answering with negated captions.

    But the researchers warning that extra work is required to deal with the foundation causes of this drawback. They hope their analysis alerts potential customers to a beforehand unnoticed shortcoming that would have severe implications in high-stakes settings the place these models are at the moment getting used, from figuring out which sufferers obtain sure therapies to figuring out product defects in manufacturing vegetation.

    “This is a technical paper, but there are bigger issues to consider. If something as fundamental as negation is broken, we shouldn’t be using large vision/language models in many of the ways we are using them now — without intensive evaluation,” says senior writer Marzyeh Ghassemi, an affiliate professor within the Department of Electrical Engineering and Computer Science (EECS) and a member of the Institute of Medical Engineering Sciences and the Laboratory for Information and Decision Systems.

    Ghassemi and Alhamoud are joined on the paper by Shaden Alshammari, an MIT graduate scholar; Yonglong Tian of OpenAI; Guohao Li, a former postdoc at Oxford University; Philip H.S. Torr, a professor at Oxford; and Yoon Kim, an assistant professor of EECS and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT. The analysis will likely be introduced at Conference on Computer Vision and Pattern Recognition.

    Neglecting negation

    Vision-language models (VLM) are skilled utilizing enormous collections of photographs and corresponding captions, which they study to encode as units of numbers, known as vector representations. The models use these vectors to differentiate between completely different photographs.

    A VLM makes use of two separate encoders, one for textual content and one for photographs, and the encoders study to output comparable vectors for a picture and its corresponding textual content caption.

    “The captions express what is in the images — they are a positive label. And that is actually the whole problem. No one looks at an image of a dog jumping over a fence and captions it by saying ‘a dog jumping over a fence, with no helicopters,’” Ghassemi says.

    Because the image-caption datasets don’t include examples of negation, VLMs by no means study to determine it.

    To dig deeper into this drawback, the researchers designed two benchmark duties that take a look at the flexibility of VLMs to know negation.

    For the primary, they used a big language mannequin (LLM) to re-caption photographs in an present dataset by asking the LLM to consider associated objects not in a picture and write them into the caption. Then they examined models by prompting them with negation words to retrieve photographs that include sure objects, however not others.

    For the second job, they designed a number of selection questions that ask a VLM to pick essentially the most acceptable caption from an inventory of carefully associated choices. These captions differ solely by including a reference to an object that doesn’t seem within the picture or negating an object that does seem within the picture.

    The models typically failed at each duties, with picture retrieval efficiency dropping by practically 25 % with negated captions. When it got here to answering a number of selection questions, the most effective models solely achieved about 39 % accuracy, with a number of models acting at and even under random likelihood.

    One purpose for this failure is a shortcut the researchers name affirmation bias — VLMs ignore negation words and give attention to objects within the photographs as an alternative.

    “This does not just happen for words like ‘no’ and ‘not.’ Regardless of how you express negation or exclusion, the models will simply ignore it,” Alhamoud says.

    This was constant throughout each VLM they examined.

    “A solvable problem”

    Since VLMs aren’t sometimes skilled on picture captions with negation, the researchers developed datasets with negation words as a primary step towards fixing the issue.

    Using a dataset with 10 million image-text caption pairs, they prompted an LLM to suggest associated captions that specify what’s excluded from the photographs, yielding new captions with negation words.

    They needed to be particularly cautious that these artificial captions nonetheless learn naturally, or it may trigger a VLM to fail in the actual world when confronted with extra advanced captions written by people.

    They discovered that finetuning VLMs with their dataset led to efficiency beneficial properties throughout the board. It improved models’ picture retrieval skills by about 10 %, whereas additionally boosting efficiency within the multiple-choice query answering job by about 30 %.

    “But our solution is not perfect. We are just recaptioning datasets, a form of data augmentation. We haven’t even touched how these models work, but we hope this is a signal that this is a solvable problem and others can take our solution and improve it,” Alhamoud says.

    At the identical time, he hopes their work encourages extra customers to consider the issue they need to use a VLM to unravel and design some examples to check it earlier than deployment.

    In the longer term, the researchers may broaden upon this work by educating VLMs to course of textual content and pictures individually, which can enhance their skill to know negation. In addition, they may develop further datasets that embody image-caption pairs for particular purposes, equivalent to well being care.

    Share. Facebook Twitter Pinterest LinkedIn WhatsApp

    Related Posts

    AI

    How a new type of AI is helping police skirt facial recognition bans

    AI

    Hybrid AI model crafts smooth, high-quality videos in seconds | Ztoog

    AI

    How to build a better AI benchmark

    AI

    Q&A: A roadmap for revolutionizing health care through data-driven innovation | Ztoog

    AI

    This data set helps researchers spot harmful stereotypes in LLMs

    AI

    Making AI models more trustworthy for high-stakes settings | Ztoog

    AI

    The AI Hype Index: AI agent cyberattacks, racing robots, and musical models

    The Future

    Meta says its Llama AI models have been downloaded 1.2B times

    Leave A Reply Cancel Reply

    Follow Us
    • Facebook
    • Twitter
    • Pinterest
    • Instagram
    Top Posts
    Crypto

    Ethereum Bearish Falling Wedge Pattern Appears, How Low Can Price Go?

    Ethereum has adopted the overall pattern of Bitcoin over the previous couple of weeks and…

    AI

    Symposium highlights scale of mental health crisis and novel methods of diagnosis and treatment | Ztoog

    Digital applied sciences, akin to smartphones and machine studying, have revolutionized schooling. At the McGovern…

    Crypto

    First French Crypto License Given to Societe Generale’s Crypto Branch

    Share this text Société Générale’s SG Forge division earns the first-ever crypto license from the…

    Crypto

    Bitcoin To Reach $1 Million In Days To Weeks, Crypto Analyst

    The CEO of Jan3 and Bitcoiner, Samson Mow, has as soon as once more reiterated…

    Mobile

    Beats Fit Pro vs Beats Studio Buds Plus: What’s the difference?

    While Beats might be most well-known for its over-the-ear headphones, it’s nonetheless a drive to…

    Our Picks
    The Future

    3-Legged Lion Sets Swimming Record in Crocodile-Filled River

    Gadgets

    Your Next Samsung OLED TV May Come With An LG Panel Thanks To A New Agreement

    Science

    Summer solstice June 2023: What is it and what causes it?

    Categories
    • AI (1,485)
    • Crypto (1,746)
    • Gadgets (1,798)
    • Mobile (1,841)
    • Science (1,856)
    • Technology (1,792)
    • The Future (1,638)
    Most Popular
    The Future

    Fortnite’s new ‘OG’ season is a trip through the game’s history

    AI

    Integrating humans with AI in structural design | Ztoog

    Crypto

    Ethereum Faces Market Tremors As Celsius Offloads $1 Billion in ETH

    Ztoog
    Facebook X (Twitter) Instagram Pinterest
    • Home
    • About Us
    • Contact us
    • Privacy Policy
    • Terms & Conditions
    © 2025 Ztoog.

    Type above and press Enter to search. Press Esc to cancel.