Graph Convolutional Networks (GCNs) are promising deep learning approaches in learning representations for graph-structured data.

The Metropolitan police's claims that their use of live facial recognition is bias-free are not substantiated by the report they cite to support their case, a leading expert on the technology has said. The Met is planning its biggest and most high profile use of LFR yet this bank holiday weekend at Notting Hill carnival in west London. The Guardian understands it will be deployed at two sites on the approaches to the carnival, with the force insisting on its use despite the Equality and Human Rights Commission saying police use of LFR is unlawful. The new claims come from Prof Pete Fussey, who led the only independent academic review of police use of facial recognition, is a former reviewer of LFR for the Met from 2018-19, and currently advises other forces in the UK and abroad on its use. The Met says it has reformed its use of LFR after a 2023 study it commissioned from the National Physical Laboratory (NPL) and it is now, in effect, bias-free. But Fussey said: "The claims the Met are making about the absence of bias from the NPL report are not substantiated by the facts in that report."

Do the main claims made in the abstract and introduction accurately reflect the paper's If you ran experiments... (a) Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Y es] (b) Did you specify all the training details (e.g., data splits, hyperparameters, how they Did you report error bars (e.g., with respect to the random seed after running experiments multiple times)? Did you include the total amount of compute and the type of resources used (e.g., type Did you mention the license of the assets? Did you include any new assets either in the supplemental material or as a URL? [Y es] Did you discuss whether and how consent was obtained from people whose data you're If you used crowdsourcing or conducted research with human subjects... (a) Any of these applications may have a different social effect. We briefly introduce the SOT A defense efforts [67, 55, 5, 69, 63, 56, 23] for GCNs here. The most related work to this paper is GCN-SVD.

Graph Convolutional Networks (GCNs) are promising deep learning approaches in learning representations for graph-structured data.

The Met commissioner should scrap plans to deploy live facial recognition (LFR) at next weekend's Notting Hill carnival because the technology is riven with "racial bias" and subject to a legal challenge, 11 civil liberty and anti-racist groups have demanded. A letter sent to Mark Rowley warns that use of instant face-matching cameras at an event that celebrates the African-Caribbean community "will only exacerbate concerns about abuses of state power and racial discrimination within your force". The Runnymede Trust, Liberty, Big Brother Watch, Race on the Agenda, and Human Rights Watch are among those who claim the technology "is less accurate for women and people of colour". The demand comes just days after ministers ramped up the deployment of vans fixed with facial recognition technology to nine forces across England and Wales. The Met said last month it would deploy specially mounted cameras at entries and exits of the two-day event in west London.

Large Language Model (LLM) pretraining traditionally relies on autoregressive language modeling on randomly sampled data blocks from web-scale datasets. We take inspiration from human learning techniques like spaced repetition to hypothesize that random data sampling for LLMs leads to high training cost and low quality models which tend to forget data. In order to effectively commit web-scale information to long-term memory, we propose the LFR (Learn, Focus, and Review) pedagogy, a new dynamic training paradigm which focuses and repeatedly reviews complex data blocks at systematic intervals based on the model's learning pace and progress. LFR records the model perplexities for different data blocks and frequently revisits blocks with higher perplexity which are more likely to be forgotten. We pretrain the GPT-2 models (124M - 1.5B) from scratch on the OpenWebText dataset using LFR. We test on downstream tasks from the language modeling, question answering, translation, and problem solving domains to achieve consistently lower perplexity and higher accuracy than the baseline OpenAI models, while obtaining a 20x pretraining speed-up.

Generating safety-critical scenarios, which are essential yet difficult to collect at scale, offers an effective method to evaluate the robustness of autonomous vehicles (AVs). Existing methods focus on optimizing adversariality while preserving the naturalness of scenarios, aiming to achieve a balance through data-driven approaches. However, without an appropriate upper bound for adversariality, the scenarios might exhibit excessive adversariality, potentially leading to unavoidable collisions. In this paper, we introduce FREA, a novel safety-critical scenarios generation method that incorporates the Largest Feasible Region (LFR) of AV as guidance to ensure the reasonableness of the adversarial scenarios. Concretely, FREA initially pre-calculates the LFR of AV from offline datasets. Subsequently, it learns a reasonable adversarial policy that controls critical background vehicles (CBVs) in the scene to generate adversarial yet AV-feasible scenarios by maximizing a novel feasibility-dependent objective function. Extensive experiments illustrate that FREA can effectively generate safety-critical scenarios, yielding considerable near-miss events while ensuring AV's feasibility. Generalization analysis also confirms the robustness of FREA in AV testing across various surrogate AV methods and traffic environments.

Parameter-efficient fine-tuning (PEFT) enables efficient adaptation of pre-trained language models (PLMs) to specific tasks. By tuning only a minimal set of (extra) parameters, PEFT achieves performance comparable to full fine-tuning. However, despite its prevalent use, the security implications of PEFT remain largely unexplored. In this paper, we conduct a pilot study revealing that PEFT exhibits unique vulnerability to trojan attacks. Specifically, we present PETA, a novel attack that accounts for downstream adaptation through bilevel optimization: the upper-level objective embeds the backdoor into a PLM while the lower-level objective simulates PEFT to retain the PLM's task-specific performance. With extensive evaluation across a variety of downstream tasks and trigger designs, we demonstrate PETA's effectiveness in terms of both attack success rate and unaffected clean accuracy, even after the victim user performs PEFT over the backdoored PLM using untainted data. Moreover, we empirically provide possible explanations for PETA's efficacy: the bilevel optimization inherently 'orthogonalizes' the backdoor and PEFT modules, thereby retaining the backdoor throughout PEFT. Based on this insight, we explore a simple defense that omits PEFT in selected layers of the backdoored PLM and unfreezes a subset of these layers' parameters, which is shown to effectively neutralize PETA.

A Beyoncé gig, the coronation of King Charles, and the British Formula One Grand Prix all have one thing in common: Thousands of people at the events, which all took place earlier this year, had their faces scanned by police-operated face recognition tech. Backed by the Conservative government, police forces across England and Wales are being told to rapidly expand their use of the highly controversial technology, which globally has led to false arrests, misidentifications, and lives derailed. Police have been told to double their use of face searches against databases by early next year--45 million passport photos could be opened up to searches--and police are increasingly working with stores to try to identify shoplifters. Simultaneously, more regional police forces are testing real-time systems in public places. The rapid expansion of face recognition comes at a time when trust in policing levels are at record lows, following a series of high-profile scandals.