We use cookies to analyze the browsing and usage of our website and to personalize your experience. You can disable these technologies at any time, but this may limit certain functionalities of the site. Read our Privacy Policy for more information.
Setting cookies
You can enable and disable the types of cookies you wish to accept. However certain choices you make could affect the services offered on our sites (e.g. suggestions, personalised ads, etc.).
Essential cookies
These cookies are necessary for the operation of the site and cannot be deactivated. (Still active)
Analytics cookies
Do you accept the use of cookies to measure the audience of our sites?
Multimedia Player
Do you accept the use of cookies to display and allow you to watch the video content hosted by our partners (YouTube, etc.)?
Publications
Capture the Flag: Uncovering Data Insights with Large Language Models
The 5-choice serial reaction time task (5-CSRTT) is commonly used to assess attention in rodents. Manipulation of this task by decreasing th… (see more)e light stimulus duration is often used to probe attentional capacity and causes a decrease in accuracy and an increase in omissions. However, although a decrease in response accuracy is commonly interpreted as a decrease in attention, it is more difficult to interpret an increase in omissions in terms of attentional performance. Here we present a series of experiments in rats that seeks to investigate the origins of these key behavioral measures of attention in the 5-CSRTT. After an initial training in the 5-CSRTT, rats were tested in a variable stimulus duration procedure to increase task difficulty and probe visual attentional capacity under several specific controlled conditions. We found that response accuracy reflects visuospatial sustained attentional processing, as commonly interpreted, while response omission reflects rats’ ignorance about the stimulus location, presumably due to failure to pay attention to the curved wall during its presentation. Moreover, when rats lack of relevant information, they choose not to respond instead of responding randomly. Overall, our results indicate that response accuracy and response omission thus correspond to two distinct attentional states.
Understanding computations in the visual system requires a characterization of the distinct feature preferences of neurons in different visu… (see more)al cortical areas. However, we know little about how feature preferences of neurons within a given area relate to that area’s role within the global organization of visual cortex. To address this, we recorded from thousands of neurons across six visual cortical areas in mouse and leveraged generative AI methods combined with closed-loop neuronal recordings to identify each neuron’s visual feature preference. First, we discovered that the mouse’s visual system is globally organized to encode features in a manner invariant to the types of image transformations induced by self-motion. Second, we found differences in the visual feature preferences of each area and that these differences generalized across animals. Finally, we observed that a given area’s collection of preferred stimuli (‘own-stimuli’) drive neurons from the same area more effectively through their dynamic range compared to preferred stimuli from other areas (‘other-stimuli’). As a result, feature preferences of neurons within an area are organized to maximally encode differences among own-stimuli while remaining insensitive to differences among other-stimuli. These results reveal how visual areas work together to efficiently encode information about the external world.
Likelihood analysis is typically limited to normally distributed noise due to the difficulty of determining the probability density function… (see more) of complex, high-dimensional, non-Gaussian, and anisotropic noise. This work presents Score-based LIkelihood Characterization (SLIC), a framework that resolves this issue by building a data-driven noise model using a set of noise realizations from observations. We show that the approach produces unbiased and precise likelihoods even in the presence of highly non-Gaussian correlated and spatially varying noise.
We use diffusion generative models to estimate the gradient of the probability density of noise with respect to data elements. In combination with the Jacobian of the physical model of the signal, we use Langevin sampling to produce independent samples from the unbiased likelihood. We demonstrate the effectiveness of the method using real data from the Hubble Space Telescope and James Webb Space Telescope.
Fine-tuning large pre-trained models has become the *de facto* strategy for developing models that are safe to deploy. However, there has be… (see more)en little work that explains how fine-tuning alters the underlying capabilities learnt by a model during pretraining: does fine-tuning yield entirely novel capabilities or does it just modulate existing ones?
We address this question empirically in *synthetic* settings with mechanistic interpretability tools (e.g., network pruning and probing) to understand how the model's underlying capabilities are changing.
Our extensive analysis of the effects of fine-tuning shows: (i) fine-tuning rarely alters the underlying model capabilities; (ii) a minimal transformation, which we call a 'wrapper', is typically learned on top of the underlying model capabilities; and (iii) further fine-tuning on a task where such wrapped capabilities are relevant leads to sample-efficient "revival'' of the capability, i.e., the model begins reusing this capability in a few gradient steps. *This indicates practitioners can unintentionally remove a model's safety wrapper by merely fine-tuning it on a superficially unrelated task.* We additionally perform analysis on language models trained on the TinyStories dataset to support our claims in a more realistic setup.
