Research in Neuroscience, as in many scientific disciplines, is undergoing a renaissance based on deep learning. Unique to Neuroscience, deep learning models can be used not only as a tool but interpreted as models of the brain. The central claims of recent deep learning-based models of brain circuits are that they make novel predictions about neural phenomena or shed light on the fundamental functions being optimized. We show, through the case-study of grid cells in the entorhinal-hippocampal circuit, that one may get neither. We begin by reviewing the principles of grid cell mechanism and function obtained from first-principles modeling efforts, then rigorously examine the claims of deep learning models of grid cells. Using large-scale architectural and hyperparameter sweeps and theory-driven experimentation, we demonstrate that the results of such models may be more strongly driven by particular, non-fundamental, and post-hoc implementation choices than fundamental truths about neural circuits or the loss function(s) they might optimize. We discuss why these models cannot be expected to produce accurate models of the brain without the addition of substantial amounts of inductive bias, an informal No Free Lunch result for Neuroscience. Based on first principles work, we provide hypotheses for what additional loss functions will produce grid cells more robustly. In conclusion, circumspection and transparency, together with biological knowledge, are warranted in building and interpreting deep learning models in Neuroscience.

Large Language Models (LLMs) inherit societal biases from their training data, potentially leading to harmful or unfair outputs. While various techniques aim to mitigate these biases, their effects are often evaluated only along the dimension of the bias being targeted. This work investigates the cross-category consequences of targeted bias mitigation. We study four bias mitigation techniques applied across ten models from seven model families, and we explore racial, religious, profession- and gender-related biases. We measure the impact of debiasing on model coherence and stereotypical preference using the StereoSet benchmark. Our results consistently show that while targeted mitigation can sometimes reduce bias in the intended dimension, it frequently leads to unintended and often negative consequences in others, such as increasing model bias and decreasing general coherence. These findings underscore the critical need for robust, multi-dimensional evaluation tools when examining and developing bias mitigation strategies to avoid inadvertently shifting or worsening bias along untargeted axes.

Reinforcement learning has emerged as a powerful paradigm for post-training large language models (LLMs) to improve reasoning. Approaches like Reinforcement Learning from Human Feedback (RLHF) and Reinforcement Learning with Verifiable Rewards (RLVR) have shown strong results, but they require extensive external supervision. We investigate an alternative class of methods, Reinforcement Learning from Internal Feedback (RLIF), which relies solely on intrinsic model-derived signals instead of external rewards. In particular, we leverage unsupervised reward proxies such as token-level entropy, trajectory-level entropy, and self-certainty. Our theoretical analysis shows these internal objectives are partially equivalent, and we empirically evaluate various RLIF strategies on challenging math reasoning benchmarks. Experimental results demonstrate that RLIF can boost the reasoning performance of base LLMs at the beginning phase of the training, matching or surpassing RLVR techniques on these tasks. However, when training progresses, performance degrades even below the model before training. Moreover, we find that RLIF yields little improvement for instruction-tuned models, indicating diminishing returns of intrinsic feedback once an LLM is already instruction-tuned. We further analyze this limitation by mixing model weights and explain the reason of RLIF's training behaviors, providing practical guidelines for integrating internal feedback signals into LLM training. We hope our analysis of internal feedback will inform more principled and effective strategies for LLM post-training.

Advances in generative models have made it possible for AI-generated text, code, and images to mirror human-generated content in many applications. Watermarking, a technique that aims to embed information in the output of a model to verify its source, is useful for mitigating the misuse of such AI-generated content. However, we show that common design choices in LLM watermarking schemes make the resulting systems surprisingly susceptible to attack---leading to fundamental trade-offs in robustness, utility, and usability. To navigate these trade-offs, we rigorously study a set of simple yet effective attacks on common watermarking systems, and propose guidelines and defenses for LLM watermarking in practice.

Whole slide image (WSI) analysis is gaining prominence within the medical imaging field. Recent advances in pathology foundation models have shown the potential to extract powerful feature representations from WSIs for downstream tasks. However, these foundation models are usually designed for general-purpose pathology image analysis and may not be optimal for specific downstream tasks or cancer types. In this work, we present Concept Anchor-guided Task-specific Feature Enhancement (CATE), an adaptable paradigm that can boost the expressivity and discriminativeness of pathology foundation models for specific downstream tasks. Based on a set of task-specific concepts derived from the pathology vision-language model with expert-designed prompts, we introduce two interconnected modules to dynamically calibrate the generic image features extracted by foundation models for certain tasks or cancer types.

Dyna-style off-policy model-based reinforcement learning (DMBRL) algorithms are a family of techniques for generating synthetic state transition data and thereby enhancing the sample efficiency of off-policy RL algorithms. This paper identifies and investigates a surprising performance gap observed when applying DMBRL algorithms across different benchmark environments with proprioceptive observations. We show that, while DMBRL algorithms perform well in OpenAI Gym, their performance can drop significantly in DeepMind Control Suite (DMC), even though these settings offer similar tasks and identical physics backends. Modern techniques designed to address several key issues that arise in these settings do not provide a consistent improvement across all environments, and overall our results show that adding synthetic rollouts to the training process -- the backbone of Dyna-style algorithms -- significantly degrades performance across most DMC environments. Our findings contribute to a deeper understanding of several fundamental challenges in model-based RL and show that, like many optimization fields, there is no free lunch when evaluating performance across diverse benchmarks in RL.

This research applies Harold Demsetz's concept of the nirvana approach to the realm of AI governance and debunks three common fallacies in various AI policy proposals--"the grass is always greener on the other side," "free lunch," and "the people could be different." Through this, I expose fundamental flaws in the current AI regulatory proposal. First, some commentators intuitively believe that people are more reliable than machines and that government works better in risk control than companies' self-regulation, but they do not fully compare the differences between the status quo and the proposed replacements. Second, when proposing some regulatory tools, some policymakers and researchers do not realize and even gloss over the fact that harms and costs are also inherent in their proposals. Third, some policy proposals are initiated based on a false comparison between the AI-driven world, where AI does lead to some risks, and an entirely idealized world, where no risk exists at all. However, the appropriate approach is to compare the world where AI causes risks to the real world where risks are everywhere, but people can live well with these risks. The prevalence of these fallacies in AI governance underscores a broader issue: the tendency to idealize potential solutions without fully considering their real-world implications. This idealization can lead to regulatory proposals that are not only impractical but potentially harmful to innovation and societal progress.

The source-free cross-domain few-shot learning (CD-FSL) task aims to transfer pretrained models to target domains utilizing minimal samples, eliminating the need for source domain data. Addressing this issue requires models to have robust generalization abilities and strong feature representation, aligning with the characteristics of large-scale pretrained models. However, large-scale models tend to lose representational ability in cross-domain scenarios due to limited sample diversity. \zlh{Given the abundant diversity provided by semantic modality, this paper leverages textual modality to enhance training sample diversity with CLP model}, meanwhile improving model transfer efficiency. Specifically, we propose the SeGD-VPT framework, which is divided into two phases. The first step aims to increase feature diversity by adding diversity prompts to each support sample, thereby generating varying input and enhancing sample diversity. Furthermore, we use diversity descriptions of classes to guide semantically meaningful learning of diversity prompts, proposing random combinations and selections of texts to increase textual diversity. Additionally, deep prompt tuning is introduced to enhance the model's transfer capability. After training of the first step, support samples with different diversity prompts are input into the CLIP backbone to generate enhanced features. After generation, the second phase trains classifiers using the generated features. Extensive experimental results across several benchmarks verify our method is comparable to SOTA source-utilized models and attain the best performance under the source-free CD-FSL setting.

Research in Neuroscience, as in many scientific disciplines, is undergoing a renaissance based on deep learning. Unique to Neuroscience, deep learning models can be used not only as a tool but interpreted as models of the brain. The central claims of recent deep learning-based models of brain circuits are that they make novel predictions about neural phenomena or shed light on the fundamental functions being optimized. We show, through the case-study of grid cells in the entorhinal-hippocampal circuit, that one may get neither. We begin by reviewing the principles of grid cell mechanism and function obtained from first-principles modeling efforts, then rigorously examine the claims of deep learning models of grid cells.

Retrieval-Augmented Generation (RAG) is widely adopted for its effectiveness and cost-efficiency in mitigating hallucinations and enhancing the domain-specific generation capabilities of large language models (LLMs). However, is this effectiveness and cost-efficiency truly a free lunch? In this study, we comprehensively investigate the fairness costs associated with RAG by proposing a practical three-level threat model from the perspective of user awareness of fairness. Specifically, varying levels of user fairness awareness result in different degrees of fairness censorship on the external dataset. We examine the fairness implications of RAG using uncensored, partially censored, and fully censored datasets. Our experiments demonstrate that fairness alignment can be easily undermined through RAG without the need for fine-tuning or retraining. Even with fully censored and supposedly unbiased external datasets, RAG can lead to biased outputs. Our findings underscore the limitations of current alignment methods in the context of RAG-based LLMs and highlight the urgent need for new strategies to ensure fairness. We propose potential mitigations and call for further research to develop robust fairness safeguards in RAG-based LLMs.