Important insights towards the explainability of neural networks reside in the characteristics of their decision boundaries. In this work, we borrow tools from the field of adversarial robustness, and propose a new perspective that relates dataset features to the distance of samples to the decision boundary. This enables us to carefully tweak the position of the training samples and measure the induced changes on the boundaries of CNNs trained on large-scale vision datasets. We use this framework to reveal some intriguing properties of CNNs. Specifically, we rigorously confirm that neural networks exhibit a high invariance to non-discriminative features, and show that the decision boundaries of a DNN can only exist as long as the classifier is trained with some features that hold them together. Finally, we show that the construction of the decision boundary is extremely sensitive to small perturbations of the training samples, and that changes in certain directions can lead to sudden invariances in the orthogonal ones. This is precisely the mechanism that adversarial training uses to achieve robustness.

Large language models (LLMs) have achievedimpressive advancements across numerous disciplines, yet the critical issue of knowledge conflicts, a major source of hallucinations, has rarely been studied. While a few research explored the conflicts between the inherent knowledge of LLMs and the retrieved contextual knowledge, a comprehensive assessment of knowledge conflict in LLMs is still missing.

This work marries influence maximization (IM) with recent work on submodular optimization from samples. The work salvages some positive results from the wreckage of previous impossibility results on IM from samples, by showing that under an SBM model of community structure in graphs, positive results for IM under sampling are possible with a new algorithm (COPS) that is a new variation on other greedy algorithms for IM. It's surprising that the removal step in the COPS algorithm is sufficient from producing the improvement seen between Margl and COPS in Figure 2 (where Margl sometimes does worse than random). Overall this is a strong contribution to the IM literature. Pros: - Brings IM closer to practical contexts by studying IM under learned influence functions - Gives rigorous analysis of this problem for SBMs - Despite simplicity of SBMs, solid evaluation shows good performance on real data Cons: - The paper is very well written, but sometimes feels like it oversimplifies the literature in the service of somewhat overstating the importance of the paper.

This work proposes a training-free approach for the detection of LLMs-generated codes, mitigating the risks associated with their indiscriminate usage. To the best of our knowledge, our research is the first to investigate zero-shot detection techniques applied to code generated by advanced black-box LLMs like ChatGPT. Firstly, we find that existing training-based or zero-shot text detectors are ineffective in detecting code, likely due to the unique statistical properties found in code structures. We then modify the previous zero-shot text detection method, DetectGPT (Mitchell et al., 2023) by utilizing a surrogate white-box model to estimate the probability of the rightmost tokens, allowing us to identify code snippets generated by language models. Through extensive experiments conducted on the python codes of the CodeContest and APPS dataset, our approach demonstrates its effectiveness by achieving state-of-the-art detection results on text-davinci-003, GPT-3.5, and GPT-4 models. Moreover, our method exhibits robustness against revision attacks and generalizes well to Java codes. We also find that the smaller code language model like PolyCoder-160M performs as a universal code detector, outperforming the billion-scale counterpart. The codes will be available at https://github.com/ Xianjun-Yang/Code_detection.git

Traditional approaches to ecological modeling, based on mathematical equations, are hampered by the qualitative nature of ecological knowledge. In this article, we demonstrate that qualitative reasoning provides alternative and productive ways for ecologists to develop, organize, and implement models. We present a qualitative theory of population dynamics and use this theory to capture and simulate commonsense theories about population and community ecology. Advantages of this approach include the possibility of deriving relevant conclusions about ecological systems without numeric data; a compositional approach that enables the reusability of models representing partial behavior; the use of a rich vocabulary describing objects, situations, relations, and mechanisms of change; and the capability to provide causal interpretations of system behavior. A number of textbooks published recently (for example, Haefner [1996]; Jørgensen and Bendoricchio [2001]) show that ecological modeling is almost synonymous with mathematical model building.

Social media systems such as weblogs, photo-and link-sharing sites, wikis, and online forums are currently thought to produce up to one third of new web content. One thing that sets these "web 2.0" sites apart from traditional web pages and resources is that they are intertwined with other forms of networked data. Their standard hyperlinks are enriched by social networks, comments, trackbacks, advertisements, tags, RDF data, and metadata. We describe recent work on building systems that use models of the blogosphere to recognize spam blogs, find opinions on topics, identify communities of interest, derive trust relationships, and detect influential bloggers. Their reach and impact is significant, with tens of millions of people providing content on a regular basis around the world.

Homeless youth are prone to human immunodeficiency virus (HIV) due to their engagement in high-risk behavior such as unprotected sex, sex under influence of drugs, and so on. Many nonprofit agencies conduct interventions to educate and train a select group of homeless youth about HIV prevention and treatment practices and rely on word-of-mouth spread of information through their one single social network Previous work in strategic selection of intervention participants does not handle uncertainties in the social networks' structure and evolving network state, potentially causing significant shortcomings in spread of information. Thus, we developed PSINET, a decision-support system to aid the agencies in this task. PSINET includes the following key novelties: (1) it handles uncertainties in network structure and evolving network state; (2) it addresses these uncertainties by using POMDPs in influence maximization; and (3) it provides algorithmic advances to allow high-quality approximate solutions for such POMDPs. Simulations show that PSINET achieves around 60 percent more information spread over the current state of the art.

Could robots change the way we think? While that might seem the stuff of dark science fiction, New Zealand artificial intelligence (AI) experts say there's real fear that computer algorithms could hijack our language, and ultimately influence our views on products or politics. "I would compare the situation with the subliminal advertising that was outlawed in the 1970s," said Associate Professor Christoph Bartneck, of Canterbury University's Human Interface Technology Laboratory, or HIT Lab. "We are in a danger of repeating the exact same issue with the use of our language." Bartneck has been working in the area with colleague Jurgen Brandstetter and other experts at the New Zealand Institute of Language Brain and Behaviour and Northwestern University in the United States.

The emerging field of artificial intelligence (AI) risks provoking a public backlash as it increasingly falls into private hands, threatens people's jobs, and operates without effective oversight or regulatory control, leading experts in the technology warn. At the start of a new Guardian series on AI, experts in the field highlight the huge potential for the technology, which is already speeding up scientific and medical research, making cities run more smoothly, and making businesses more efficient. But for all the promise of an AI revolution, there are mounting social, ethical and political concerns about the technology being developed without sufficient oversight from regulators, legislators and governments. The report found that AI had the potential to add £630bn to the economy by 2035. But to reap the rewards, the technology must benefit society, she said.