Large language models (LLMs) are increasingly expected to tackle complex tasks, driven by their expanding applications and users' growing proficiency in crafting sophisticated prompts. However, as the number of explicitly stated requirements increases (particularly more than 10 constraints), LLMs often struggle to accurately follow such complex instructions, which limits their applicability in complex real-world scenarios. To the best of our knowledge, existing datasets do not exceed 10 constraints per instance. To address this challenge, we propose RECAST, an efficient and scalable framework for synthesizing datasets where each example incorporates far more constraints than those in existing benchmarks, aiming to challenge and extend the boundaries of models' ability to follow complex instructions. These constraints are extracted from real-world prompt-response pairs to ensure practical relevance. Using this framework, we construct RECAST-30K, a large-scale, high-quality dataset comprising 30k instances spanning 19 constraint types. Experimental results demonstrate that models finetuned on RECAST-30K substantially improve in following complex instructions while maintaining their general capabilities without degradation. Moreover, RECAST enables automatic verification of constraint satisfaction via rule-based validators for quantitative constraints and LLM-based validators for qualitative ones; the verifiability provided by RECAST enables the design of reward functions for reinforcement learning, which further boosts model performance on complex and challenging tasks.

Incremental learning aims to adapt to new sets of categories over time with minimal computational overhead. Prior work often addresses this task by training efficient task-specific adaptors that modify frozen layer weights or features to capture relevant information without affecting predictions on previously learned categories. While these adaptors are generally more efficient than finetuning the entire network, they still require tens to hundreds of thousands of task-specific trainable parameters even for relatively small networks, making it challenging to operate on resource-constrained environments with high communication costs like edge devices or mobile phones. Thus, we propose Reparameterized, Compact weight Adaptation for Sequential Tasks (RECAST), a novel method that dramatically reduces task-specific trainable parameters to fewer than 50 - several orders of magnitude less than competing methods like LoRA. RECAST accomplishes this efficiency by learning to decompose layer weights into a soft parameter-sharing framework consisting of shared weight templates and very few module-specific scaling factors or coefficients. This soft parameter-sharing framework allows for effective task-wise reparameterization by tuning only these coefficients while keeping templates frozen.A key innovation of RECAST is the novel weight reconstruction pipeline called Neural Mimicry, which eliminates the need for pretraining from scratch. This allows for high-fidelity emulation of existing pretrained weights within our framework and provides quick adaptability to any model scale and architecture. Extensive experiments across six datasets demonstrate RECAST outperforms the state-of-the-art by up to 3% across various scales, architectures, and parameter spaces Moreover, we show that RECAST's architecture-agnostic nature allows for seamless integration with existing methods, further boosting performance.

Untreated periodontitis causes inflammation within the supporting tissue of the teeth and can ultimately lead to tooth loss. Modeling periodontal outcomes is beneficial as they are difficult and time consuming to measure, but disparities in representation between demographic groups must be considered. There may not be enough participants to build group specific models and it can be ineffective, and even dangerous, to apply a model to participants in an underrepresented group if demographic differences were not considered during training. We propose an extension to RECaST Bayesian transfer learning framework. Our method jointly models multivariate outcomes, exhibiting significant improvement over the previous univariate RECaST method. Further, we introduce an online approach to model sequential data sets. Negative transfer is mitigated to ensure that the information shared from the other demographic groups does not negatively impact the modeling of the underrepresented participants. The Bayesian framework naturally provides uncertainty quantification on predictions. Especially important in medical applications, our method does not share data between domains. We demonstrate the effectiveness of our method in both predictive performance and uncertainty quantification on simulated data and on a database of dental records from the HealthPartners Institute.

Transfer learning uses a data model, trained to make predictions or inferences on data from one population, to make reliable predictions or inferences on data from another population. Most existing transfer learning approaches are based on fine-tuning pre-trained neural network models, and fail to provide crucial uncertainty quantification. We develop a statistical framework for model predictions based on transfer learning, called RECaST. The primary mechanism is a Cauchy random effect that recalibrates a source model to a target population; we mathematically and empirically demonstrate the validity of our RECaST approach for transfer learning between linear models, in the sense that prediction sets will achieve their nominal stated coverage, and we numerically illustrate the method's robustness to asymptotic approximations for nonlinear models. Whereas many existing techniques are built on particular source models, RECaST is agnostic to the choice of source model. For example, our RECaST transfer learning approach can be applied to a continuous or discrete data model with linear or logistic regression, deep neural network architectures, etc. Furthermore, RECaST provides uncertainty quantification for predictions, which is mostly absent in the literature. We examine our method's performance in a simulation study and in an application to real hospital data.

The most crucial facet of Artificial Intelligence (AI) is developing the technology without turning a blind eye to its consequences. AI is ultimately built by human beings, and humans can have very diverse motives for why they create something. Unfortunately, today there is a massive gap between people making these systems and those impacted by these systems. The changes that AI will bring to the jobs done by humans will have marked consequences on societies, economies and labour markets. For example, robotics has enabled us to do work that was too dangerous or impossible for humans or that is done more effectively by a machine.

George Miller's Mad Max: Fury Road may have spent almost 20 years stuck in development hell, but the filmmaker had always envisioned the project without Mel Gibson in the lead. The 1979 original was the breakthrough role of the actor's career, and the sequel remains one of the greatest action movies ever made, but Miller was adamant that the title hero be recast after the Lethal Weapon star had aged out of consideration. Tom Hardy, 22 years Gibson's junior, was chosen to play the new Max Rockatansky instead, but in terms of the franchise's canon, he's the same character, just at a different stage of his life. The choice to recast was a wise one, too, as Fury Road left audiences with their jaws on the floor when it exploded onto the scene and almost instantly gained a reputation as a modern action classic, scoring the rare combination of box office success, universal critical acclaim and awards season glory. The post-apocalyptic blockbuster earned ten Academy Award nominations including Best Picture and Best Director, and ended up walking away with six prizes in the technical categories.

In May 2015, The Simpsons voice actor Harry Shearer--who plays a number of key characters including, quite incredibly, both Mr Burns and Waylon Smithers--announced that he was leaving the show. This story originally appeared on WIRED UK. By then, the animated series had been running for more than 25 years, and the pay of its vocal cast had risen from $30,000 an episode in 1998 to $400,000 an episode from 2008 onwards. But Fox, the producer of The Simpsons, was looking to cut costs-- and was threatening to cancel the series unless the voice actors took a 30 percent pay cut. Most of them agreed, but Shearer (who had been critical of the show's declining quality) refused to sign--after more than two decades, he wanted to break out of the golden handcuffs, and win back the freedom and the time to pursue his own work.

In May 2015, The Simpsons voice actor Harry Shearer – who plays a number of key characters including, quite incredibly, both Mr Burns and Waylon Smithers – announced that he was leaving the show. By then, the animated series had been running for more than 25 years, and the pay of its vocal cast had risen from $30,000 an episode in 1998 to $400,000 an episode from 2008 onwards. But Fox, the producer of The Simpsons, was looking to cut costs – and was threatening to cancel the series unless the voice actors took a 30 per cent pay cut. Most of them agreed, but Shearer (who had been critical of the show's declining quality) refused to sign – after more than two decades, he wanted to break out of the golden handcuffs, and win back the freedom and the time to pursue his own work. Showrunner Al Jean said Shearer's iconic characters – who also include Principal Skinner, Ned Flanders and Otto Mann – would be recast.

For many applications it is critical to know the uncertainty of a neural network's predictions. While a variety of neural network parameter estimation methods have been proposed for uncertainty estimation, they have not been rigorously compared across uncertainty measures. We assess four of these parameter estimation methods to calibrate uncertainty estimation using four different uncertainty measures: entropy, mutual information, aleatoric uncertainty and epistemic uncertainty. We also evaluate their calibration using expected calibration error. We additionally propose a novel method of neural network parameter estimation called RECAST, which combines cosine annealing with warm restarts with Stochastic Gradient Langevin Dynamics, capturing more diverse parameter distributions. When benchmarked against mutilated data from MNIST, we show that RECAST is well-calibrated and when combined with predictive entropy and epistemic uncertainty it offers the best calibrated measure of uncertainty when compared to recent methods.