搜索结果：yourself

As Large Language Model (LLM) agents increasingly operate in complex environments with real-world consequences, their safety becomes critical. While uncertainty quantification is well-studied for single-turn tasks, multi-turn agentic scenarios with real-world tool access present unique challenges where uncertainties and ambiguities compound, leading to severe or catastrophic risks beyond traditional text generation failures. We propose using "quitting" as a simple yet effective behavioral mechanism for LLM agents to recognize and withdraw from situations where they lack confidence. Leveraging the ToolEmu framework, we conduct a systematic evaluation of quitting behavior across 12 state-of-the-art LLMs. Our results demonstrate a highly favorable safety-helpfulness trade-off: agents prompted to quit with explicit instructions improve safety by an average of +0.39 on a 0-3 scale across all models (+0.64 for proprietary models), while maintaining a negligible average decrease of -0.03 in helpfulness. Our analysis demonstrates that simply adding explicit quit instructions proves to be a highly effective safety mechanism that can immediately be deployed in existing agent systems, and est

Misinformation is "sticky" in nature, requiring a considerable effort to undo its influence. One such effort is debunking or exposing the falsity of information. As an abundance of misinformation is on social media, platforms do bear some debunking responsibility in order to preserve their trustworthiness as information providers. A subject of interpretation, platforms poorly meet this responsibility and allow dangerous health misinformation to influence many of their users. This open route to harm did not sit well with health professional users, who recently decided to take the debunking into their own hands. To study this individual debunking effort - which we call 'Debunk-It-Yourself (DIY)' - we conducted an exploratory survey n=14 health professionals who wage a misinformation counter-influence campaign through videos on TikTok. We focused on two topics, nutrition and mental health, which are the ones most often subjected to misinformation on the platform. Our thematic analysis reveals that the counterinfluence follows a common process of initiation, selection, creation, and "stitching" or duetting a debunking video with a misinformation video. The 'Debunk-It-Yourself' effort w

Future-Proof Yourself is a practical guide that helps readers navigate the fast-changing world of artificial intelligence in everyday life. The book begins by explaining how computers learn from data in simple, relatable terms, and gradually introduces the methods used in modern AI. It shows how basic ideas in machine learning evolve into advanced systems that can recognize images, understand language, and even make decisions. The guide also reviews the history of AI and highlights the major breakthroughs that have shaped its growth. Looking ahead, the book explores emerging trends such as the integration of AI with digital twins, wearable devices, and virtual environments. Designed for a general audience, the text avoids heavy technical jargon and presents complex ideas in clear, straightforward language so that anyone can gain a solid understanding of the technology that is set to transform our future.

Existing commercial and in-house software development tools are often inaccessible to Blind and Low Vision Software Professionals (BLVSPs), hindering their participation and career growth at work. Building on existing research on Do-It-Yourself (DIY) Assistive Technologies and customized tools made by programmers, we shed light on the currently unexplored intersection of how DIY tools built and used by BLVSPs support accessible software development. Through semi-structured interviews with 30 BLVSPs, we found that such tools serve many different purposes and are driven by motivations such as desiring to maintain a professional image and a sense of dignity at work. These tools had significant impacts on workplace accessibility and revealed a need for a more centralized community for sharing tools, tips, and tricks. Based on our findings, we introduce the "Double Hacker Dilemma" and highlight a need for developing more effective peer and organizational platforms that support DIY tool sharing.

We present UP2You, the first tuning-free solution for reconstructing high-fidelity 3D clothed portraits from extremely unconstrained in-the-wild 2D photos. Unlike previous approaches that require "clean" inputs (e.g., full-body images with minimal occlusions, or well-calibrated cross-view captures), UP2You directly processes raw, unstructured photographs, which may vary significantly in pose, viewpoint, cropping, and occlusion. Instead of compressing data into tokens for slow online text-to-3D optimization, we introduce a data rectifier paradigm that efficiently converts unconstrained inputs into clean, orthogonal multi-view images in a single forward pass within seconds, simplifying the 3D reconstruction. Central to UP2You is a pose-correlated feature aggregation module (PCFA), that selectively fuses information from multiple reference images w.r.t. target poses, enabling better identity preservation and nearly constant memory footprint, with more observations. We also introduce a perceiver-based multi-reference shape predictor, removing the need for pre-captured body templates. Extensive experiments on 4D-Dress, PuzzleIOI, and in-the-wild captures demonstrate that UP2You consiste

Multi-modal Large Language Models (MLLMs) integrate visual and linguistic reasoning to address complex tasks such as image captioning and visual question answering. While MLLMs demonstrate remarkable versatility, MLLMs appears limited performance on special applications. But tuning MLLMs for downstream tasks encounters two key challenges: Task-Expert Specialization, where distribution shifts between pre-training and target datasets constrain target performance, and Open-World Stabilization, where catastrophic forgetting erases the model general knowledge. In this work, we systematically review recent advancements in MLLM tuning methodologies, classifying them into three paradigms: (I) Selective Tuning, (II) Additive Tuning, and (III) Reparameterization Tuning. Furthermore, we benchmark these tuning strategies across popular MLLM architectures and diverse downstream tasks to establish standardized evaluation analysis and systematic tuning principles. Finally, we highlight several open challenges in this domain and propose future research directions. To facilitate ongoing progress in this rapidly evolving field, we provide a public repository that continuously tracks developments: ht

*Minimal sufficient reasons* represent a prevalent form of explanation - the smallest subset of input features which, when held constant at their corresponding values, ensure that the prediction remains unchanged. Previous *post-hoc* methods attempt to obtain such explanations but face two main limitations: (1) Obtaining these subsets poses a computational challenge, leading most scalable methods to converge towards suboptimal, less meaningful subsets; (2) These methods heavily rely on sampling out-of-distribution input assignments, potentially resulting in counterintuitive behaviors. To tackle these limitations, we propose in this work a self-supervised training approach, which we term *sufficient subset training* (SST). Using SST, we train models to generate concise sufficient reasons for their predictions as an integral part of their output. Our results indicate that our framework produces succinct and faithful subsets substantially more efficiently than competing post-hoc methods, while maintaining comparable predictive performance.

Roughly, a metric space has padding parameter $β$ if for every $Δ>0$, there is a stochastic decomposition of the metric points into clusters of diameter at most $Δ$ such that every ball of radius $γΔ$ is contained in a single cluster with probability at least $e^{-γβ}$. The padding parameter is an important characteristic of a metric space with vast algorithmic implications. In this paper we prove that the shortest path metric of every $K_r$-minor-free graph has padding parameter $O(\log r)$, which is also tight. This resolves a long standing open question, and exponentially improves the previous bound. En route to our main result, we construct sparse covers for $K_r$-minor-free graphs with improved parameters, and we prove a general reduction from sparse covers to padded decompositions.

On the surface, behavioural science and physics seem to be two disparate fields of research. However, a closer examination of problems solved by them reveals that they are uniquely related to one another. Exemplified by the theories of quantum mind, cognition and decision-making, this unique relationship serves as the topic of this chapter. Surveying the current academic journal papers and scholarly monographs, we present an alternative vision of the role of quantum mechanics in the modern studies of human perception, behaviour and decision-making. To that end, we mostly aim to answer the 'how' question, deliberately avoiding complex mathematical concepts but developing a technically simple computational code that the readers can modify to design their own quantum-inspired models. We also present several practical examples of the application of the computation code and outline several plausible scenarios, where quantum models based on the proposed do-it-yourself model kit can help understand the differences between the behaviour of individuals and social groups.

Open set recognition (OSR) is a critical aspect of machine learning, addressing the challenge of detecting novel classes during inference. Within the realm of deep learning, neural classifiers trained on a closed set of data typically struggle to identify novel classes, leading to erroneous predictions. To address this issue, various heuristic methods have been proposed, allowing models to express uncertainty by stating "I don't know." However, a gap in the literature remains, as there has been limited exploration of the underlying mechanisms of these methods. In this paper, we conduct an analysis of open set recognition methods, focusing on the aspect of feature diversity. Our research reveals a significant correlation between learning diverse discriminative features and enhancing OSR performance. Building on this insight, we propose a novel OSR approach that leverages the advantages of feature diversity. The efficacy of our method is substantiated through rigorous evaluation on a standard OSR testbench, demonstrating a substantial improvement over state-of-the-art methods.

Few-shot class-incremental learning (FSCIL) aims to learn sequential classes with limited samples in a few-shot fashion. Inherited from the classical class-incremental learning setting, the popular benchmark of FSCIL uses averaged accuracy (aAcc) and last-task averaged accuracy (lAcc) as the evaluation metrics. However, we reveal that such evaluation metrics may not provide adequate emphasis on the novel class performance, and the continual learning ability of FSCIL methods could be ignored under this benchmark. In this work, as a complement to existing metrics, we offer a new metric called generalized average accuracy (gAcc) which is designed to provide an extra equitable evaluation by incorporating different perspectives of the performance under the guidance of a parameter $α$. We also present an overall metric in the form of the area under the curve (AUC) along the $α$. Under the guidance of gAcc, we release the potential of intermediate features of the vision transformers to boost the novel-class performance. Taking information from intermediate layers which are less class-specific and more generalizable, we manage to rectify the final features, leading to a more generalizable

It is challenging to accelerate the training process while ensuring both high-quality generated voices and acceptable inference speed. In this paper, we propose a novel neural vocoder called InstructSing, which can converge much faster compared with other neural vocoders while maintaining good performance by integrating differentiable digital signal processing and adversarial training. It includes one generator and two discriminators. Specifically, the generator incorporates a harmonic-plus-noise (HN) module to produce 8kHz audio as an instructive signal. Subsequently, the HN module is connected with an extended WaveNet by an UNet-based module, which transforms the output of the HN module to a latent variable sequence containing essential periodic and aperiodic information. In addition to the latent sequence, the extended WaveNet also takes the mel-spectrogram as input to generate 48kHz high-fidelity singing voices. In terms of discriminators, we combine a multi-period discriminator, as originally proposed in HiFiGAN, with a multi-resolution multi-band STFT discriminator. Notably, InstructSing achieves comparable voice quality to other neural vocoders but with only one-tenth of the

Multimodal Large Language Model (MLLM) have demonstrated strong generalization capabilities across diverse distributions and tasks, largely due to extensive pre-training datasets. Fine-tuning MLLM has become a common practice to improve performance on specific downstream tasks. However, during fine-tuning, MLLM often faces the risk of forgetting knowledge acquired during pre-training, which can result in a decline in generalization abilities. To balance the trade-off between generalization and specialization, we propose measuring the parameter importance for both pre-trained and fine-tuning distributions, based on frozen pre-trained weight magnitude and accumulated fine-tuning gradient values. We further apply an importance-aware weight allocation strategy, selectively updating relatively important parameters for downstream tasks. We conduct empirical evaluations on both image captioning and visual question-answering tasks using various MLLM architectures. The comprehensive experimental analysis demonstrates the effectiveness of the proposed solution, highlighting the efficiency of the crucial modules in enhancing downstream specialization performance while mitigating generalizatio

With direct access to human-written reference as memory, retrieval-augmented generation has achieved much progress in a wide range of text generation tasks. Since better memory would typically prompt better generation~(we define this as primal problem). The traditional approach for memory retrieval involves selecting memory that exhibits the highest similarity to the input. However, this method is constrained by the quality of the fixed corpus from which memory is retrieved. In this paper, by exploring the duality of the primal problem: better generation also prompts better memory, we propose a novel framework, selfmem, which addresses this limitation by iteratively employing a retrieval-augmented generator to create an unbounded memory pool and using a memory selector to choose one output as memory for the subsequent generation round. This enables the model to leverage its own output, referred to as self-memory, for improved generation. We evaluate the effectiveness of selfmem on three distinct text generation tasks: neural machine translation, abstractive text summarization, and dialogue generation, under two generation paradigms: fine-tuned small model and few-shot LLM. Our appr

Modern object detectors are vulnerable to adversarial examples, which may bring risks to real-world applications. The sparse attack is an important task which, compared with the popular adversarial perturbation on the whole image, needs to select the potential pixels that is generally regularized by an $\ell_0$-norm constraint, and simultaneously optimize the corresponding texture. The non-differentiability of $\ell_0$ norm brings challenges and many works on attacking object detection adopted manually-designed patterns to address them, which are meaningless and independent of objects, and therefore lead to relatively poor attack performance. In this paper, we propose Adversarial Semantic Contour (ASC), an MAP estimate of a Bayesian formulation of sparse attack with a deceived prior of object contour. The object contour prior effectively reduces the search space of pixel selection and improves the attack by introducing more semantic bias. Extensive experiments demonstrate that ASC can corrupt the prediction of 9 modern detectors with different architectures (\e.g., one-stage, two-stage and Transformer) by modifying fewer than 5\% of the pixels of the object area in COCO in white-bo

The human visual system is remarkable in learning new visual concepts from just a few examples. This is precisely the goal behind few-shot class incremental learning (FSCIL), where the emphasis is additionally placed on ensuring the model does not suffer from "forgetting". In this paper, we push the boundary further for FSCIL by addressing two key questions that bottleneck its ubiquitous application (i) can the model learn from diverse modalities other than just photo (as humans do), and (ii) what if photos are not readily accessible (due to ethical and privacy constraints). Our key innovation lies in advocating the use of sketches as a new modality for class support. The product is a "Doodle It Yourself" (DIY) FSCIL framework where the users can freely sketch a few examples of a novel class for the model to learn to recognize photos of that class. For that, we present a framework that infuses (i) gradient consensus for domain invariant learning, (ii) knowledge distillation for preserving old class information, and (iii) graph attention networks for message passing between old and novel classes. We experimentally show that sketches are better class support than text in the context