Illustration by Avishek Das/SOPA Images/LightRocket via Getty Images Wiz was acquired by Google this year. A talent leader said Wiz has since seen a jump in job apps. After being bought by a large company, late-stage ventures can offer more resources than younger startups. What startups value most in candidates tends to shift as they mature, the talent leader said. A blockbuster acquisition, IPO, or funding round can turn a startup into a hiring magnet. It can also change what it takes to land a job there. Earlier this year, Google completed its $32 billion acquisition of cybersecurity startup Wiz, marking the search giant's largest deal to date. Erin Gard, vice president of talent at Wiz, told Business Insider that the company has seen a significant jump in job applications since the deal was announced a little over a year ago. She also said Wiz, which has maintained its brand since the acquisition, is hiring for more than 100 open roles across sales, engineering, and field operations. "Startups are places that a lot of people want to be in, particularly right now, so our applications are quite high," said Gard. Early-stage startups can offer big upside before a major liquidity event or financing milestone, as Business Insider recently reported. Later-stage ventures have their own draw: more resources, greater job security, and additional room for advancement. Stiff competition isn't the only hurdle to getting a foot in the door, though. Gard said candidates need to show they're equipped to support the company's current and next stage of growth — not the scrappy startup it used to be. For job seekers hoping to join a mature venture like Wiz, she shared the following advice: Tailor your pitch to the company's current stage The skills most prized at a startup can change as quickly as the company grows. Candidates should be ready to show how their experience aligns with the business at its current stage. When Gard started at Wiz in 2022, she said she kept tabs on candidates in a spreadsheet. Today, her job requires fluency with applicant-tracking systems that can bring order to a much larger hiring operation. "Being able to demonstrate an understanding of the specific chapter that an organization is in is really, really helpful," Gard said. "That shows you're ready for that stage of growth, and you're going to be able to plug in and help support us where we're at." Talk about the customer, not just the company At a mature startup, candidates should be careful not to frame their interest around the company's newfound status. A stronger pitch focuses on how they can help sustain its momentum and drive its next stage of growth, said Gard. They should also show they understand the specific challenges it's trying to solve and explain how their experience would help address them, she added. Gard cited a solutions engineer candidate who, when asked "Why Wiz?," described how customers at his previous employer struggled to prioritize millions of security findings. He explained how Wiz could help solve that problem and how he would support customers in the role. "He immediately knew the value proposition," Gard said, adding that the candidate's explanation of how customers would use Wiz in the real world "was spot on." The candidate was hired and has since been promoted to a management role, she added. Build relationships with insiders before a position opens Job seekers should thoughtfully engage with a startup's leaders and hiring managers on social media, such as by commenting on posts about product launches or asking insightful questions. Gard said that kind of engagement can signal genuine interest and help establish credibility before an application is submitted. That can be true no matter what stage company you're looking to join. Jesse Dwyer, Perplexity's chief communications officer and a hiring manager, previously told Business Insider that participating on a startup's social media channels can be especially effective, since these companies often have more flexibility in how they identify and engage with applicants. Don't focus only on successes Many candidates avoid discussing failures, said Gard, but she argued that hiring managers often favor those who can talk about setbacks, pivots, and lessons learned. Because startups inevitably make mistakes and adapt quickly, she said that candidates who demonstrate resilience and self-awareness tend to come across as more credible. "Being able to show that you too have gone through your own journey, similar to how a startup has, resonates really, really well," said Gard. "It's something that I think a lot of people are really scared to do." Read the original article on Business Insider

MTStock Studio/Getty Images Anthropic has restored access to Fable 5 after the White House cut foreign access last month. UK-based business owner Sean McDonnell was mid-task with the model when foreign access was blocked. He is back to using the AI model and said having backup plans for AI disruptions is key. This as-told-to essay is based on a conversation with Sean McDonnell, 43, who lives in England. McDonnell is the founder of the web design company Kaizen and the SaaS website Consigns. The following has been edited for length and clarity. Last month, a few hours into trying Anthropic's new Fable model, I was mid-task when the US government forced Anthropic to cut off foreign access to Fable 5 with little to no notice. When Fable was cut off, I was well-prepared and had a backup plan, so it didn't have a hugely disruptive impact on the web design business and the SaaS website I started earlier this year. We're a small team, and AI tools are a big help. I couldn't have made my website and other services as advanced as they are without AI. However, this incident was a big reminder that you absolutely need a plan B — and even a plan C — to fall back on when working with AI. The mistake many people make is to depend on AI wholly. When Fable came back, we were already prepping for it to disappear again General access to Fable has been restored, and we've already been using it. This time, the first thing we did was to make sure that everything across the codebase is saved as a Claude Skill. That way, even if we lose access to the model again, other models can reference that skill to get a good map of the codebase from the more advanced model. It's important to keep records that exist outside of the AI tool and make sure things are documented as they're developed. If Claude knows all about our code base, but it gets pulled tomorrow, would I be able to give the details over to a developer? At this stage, I think I could, because I've been documenting everything as I go. It's a fail-safe. The key is to augment, not replace, our usual workflows with AI. The first time Fable went down, I had a different backup plan in place In the past, when we used Claude's Opus 4.6, it would stop mid-task because it hit the token limit so quickly. We didn't realize how token-heavy the tool was, and it left our codebase in a bit of a mess. Since we'd learned this lesson with 4.6, we were more prepared for unforeseen circumstances when we started using Fable for the first time last month. Last month, before foreign access was cut off, I asked Fable to create a guide that both Claude and other AI models could follow. This enabled us to pass the remaining tasks to other agents when we lost access to Fable. We passed some to Codex and others to Claude 4.8. If we hadn't been prepared this way, the Fable issue could've resulted in lots of work being out the window. I hope that if this happens again, there will be better communication from the company More notice would've been very useful, but I can understand why there was none if there's a matter of "security" at play. However, Anthropic, OpenAI, Gemini, and other AI companies, as well as businesses that rely on AI, including small companies like ours, should always have backup plans and be prepared for disruptions. It would've been nice to have a smoother transition and not just have a model yanked offline. Even some messaging explaining the situation within their own apps wouldn't go amiss. Do you have a similar story to share? If so, please reach out to the reporter at aapplegate@businessinsider.com. Read the original article on Business Insider

PLUS: Delegate team tasks to Claude inside Slack

As the parent of two little girls, I often think about how their childhood is different from mine. The seven-year-old is learning about AI at school. The five-year-old is given internet-based homework every week. And they are both absolutely repulsed by the idea of smoking. That was not the prevailing sentiment when I was young.…

Janko Roettgers / The Verge: A look back at BitTorrent, launched by Bram Cohen 25 years ago, and how media piracy fueled its growth while its architecture shielded it from legal liability  —  Cath Virginia / The Verge … Twenty-five years ago today, a young, little-known programmer by the name of Bram Cohen fired off …

Financial Times: Sources: Anthropic moves to close loopholes that let Chinese firms like Ant use its models via workarounds including cloud providers and overseas subsidiaries  —  Engineers are still finding ways to use AI models despite stringent restrictions.  Anthropic is moving to shut loopholes …

A government customer of NSO Group used the company's Pegasus spyware to hack into the phone of a European politician, who at the time was serving on an EU committee tasked with investigating the spyware industry.

Discussion | Link

Dawn Lim / Bloomberg: Blackstone's QTS abandons plans to build its portion of a 2,100-acre data center campus in Virginia, following years of local opposition and legal challenges  —  Blackstone Inc.'s QTS is walking away from plans to build its portion of a 2,100-acre data center campus in Virginia …

arXiv:2607.01312v1 Announce Type: new Abstract: Visual narratives are central to storyboards, comics, children's media, and film previsualization, where viewers understand stories from images alone. Recent generators such as StoryDiffusion produce coherent sequences, but visual coherence does not guarantee that source-story transition meaning remains recoverable. Existing benchmarks assess visual quality, content faithfulness, and scene coherence, but miss a critical failure mode: storyboards where scenes appear visually coherent while the semantic link between scenes disappears. We introduce KathaTrace, a generator-agnostic protocol for diagnosing semantic trajectory collapse, defined as the loss of transition meaning needed to understand how one scene follows another. KathaTrace evaluates transitions under three evidence conditions: text-only, image-only, and text-plus-image, and filters ambiguous items. We contribute KathaBench-25K, with 5,000 narratives from classical collections including Aesop, Panchatantra, and Kathasaritasagara, 20,000 transitions, and 28,712 recoverability questions. We define Semantic Trajectory Gap, or STG, as text-only minus image-only recoverability, measuring transition meaning lost during visualization. Human validation yields Fleiss' kappa = 0.845. Experiments across state-of-the-art generators show substantial STG of 23.5 +/- 1.3. Semantic Compass, an actionability probe, uses KathaTrace signals for post-generation repair and improves storyboard selection.

arXiv:2607.01353v1 Announce Type: new Abstract: Video-based person re-identification (Re-ID) aims to retrieve the same identity in the query video clips from the gallery video clips. To solve this problem, exploiting fine-grained features is of great importance, especially when discriminating identities that are similar in appearance. In this paper, we propose to enhance the ability to explore fine-grained information with a novel input-aware extendable expert module. Instead of updating the network parameters with every sample in the dataset, we aim to train the experts within specific subsets that only contain similar samples and promote their ability to exploit fine-grained information within these similar samples. To achieve this goal, we incorporate two mechanisms in this module: input-aware expert selection mechanism and spatial-temporal selection mechanism. The first mechanism dynamically activates a set of experts on subsets of similar samples, pushing the experts to exploit subtle differences between these similar samples, while the second one further increases their sensitivity to the fine-grained differences in spatial and temporal aspects and allows the experts to dynamically utilize them for different input samples. In addition, to facilitate the expert module, we design an extendable scheme that allows the module to flexibly add new experts when necessary. As a result, our method achieves outstanding performance on two large-scale datasets.

arXiv:2607.01370v1 Announce Type: new Abstract: High definition map generation is essential for autonomous driving, yet remains a labor-intensive process at scale. We present MapDreamer, a generative diffusion model that synthesizes lane-level vector maps with explicit topology directly from a single aerial image. MapDreamer learns a compact latent representation of lane centerlines and their topological relations using a variational autoencoder and predicts graphs with a transformer-based latent diffusion model. To align generated maps with the observed scene, we condition each denoising step on dense aerial features injected through cross-attention. To handle the varying number of lanes across scenes, we propose a lane cardinality module paired with background ghost lane latents, a learned buffer that prevents slot collapse during diffusion. Furthermore, we introduce a sliding-window global graph aggregation strategy that stitches local tiles into city-scale maps while preserving connectivity through encoded lane boundaries. Experiments on UrbanLaneGraph derived from Argoverse 2 show improved geometric and topological fidelity over non-generative baselines.

arXiv:2607.01383v1 Announce Type: new Abstract: Multi-subject personalized image generation requires the precise rendering of all requested reference identities and their specified interactions based on a guiding prompt. However, state-of-the-art models still struggle with this process, frequently omitting subjects, failing to preserve reference appearances, or misattributing interactions. Furthermore, existing metrics designed primarily for single-subject fidelity cannot reliably capture these errors, suffering severe degradation in ranking separability and failing to align with human preference as the subject count increases. To address this gap, we introduce Multi-subject Interaction Benchmark and Evaluator (MIBE), a unified framework comprising a Multi-subject Interaction Benchmark (MIB) and a Multi-subject Interaction Evaluator (MIE). MIB systematically covers diverse relation types and scene complexities through a decoupled data regime. This consists of a 60K-pair VLM-labeled Silver Set for scalable metric training and a 4K-pair double-blind Human Evaluation Gold Set covering a diverse range of state-of-the-art generators, with the Silver Set reaching 95.1% cross-VLM preference agreement. To demonstrate the utility of this benchmark, we present MIE, a lightweight, reference-conditioned evaluator trained exclusively on the Silver Set with a dual-head ranking and diagnosis objective. MIE exhibits strong cross-generator generalization on the Gold Set, achieving 0.922 overall pairwise accuracy against human preference, including 0.982 on seen generators and 0.884 on unseen generators. By outperforming a broad spectrum of baseline metrics, including CLIP and DINO variants, MIE demonstrates that diagnostic supervision can preserve ranking separability and human alignment where traditional evaluators collapse.

arXiv:2607.01395v1 Announce Type: new Abstract: At the heart of human visual perception lies the ability to maintain a continuous and coherent understanding of the external world. By integrating observations with accumulated experience, the human visual system can continuously adapt to variations in both the target and its surrounding environment, while preserving robust visual continuity as scene dynamics evolve. Human vision can therefore integrate prior knowledge, spatial geometry, and semantic context to understand complex scenes and their changes. As a core problem in computer vision, visual object tracking aims to bring machine perception closer to human visual perception. These capabilities are central to the task of Generic Object Tracking (GOT). In this task, a visual tracker is initialized only with the bounding box of an arbitrarily specified target in the first frame, and must continuously localize the target in subsequent dynamic visual streams. However, future events, observations, and real-world variations are inherently unpredictable; therefore, the model's generalization and online adaptation capabilities remain bottlenecks. Tracking reliability can deteriorate when the target undergoes severe deformation, is affected by complex distractors, encounters significant environmental changes, or belongs to a category unseen during training. This dissertation aims to narrow the gap between machine visual tracking systems and human visual perception by proposing a series of methods that systematically enhance the target discrimination, robust adaptation, and geometric reasoning capabilities of tracking models.

arXiv:2607.01396v1 Announce Type: new Abstract: Urban expansion threatens global biodiversity, especially affecting arboreal species due to the fragmentation of forest habitats. The movement of arboreal species across disjointed forest patches increases mortality risk and, thus, compromises their conservation. In this context, the installation of canopy bridges can be a viable strategy; yet continuous monitoring of their use by arboreal species is essential for ensuring their effectiveness, typically carried out with the aid of camera traps. However, this method often produces false-positive images that demand time from conservationists for review. In this context, computer vision algorithms can optimize the task of detecting target species using the canopy bridges. In this study, we explored the automatic detection of brown howler monkeys (Alouatta guariba) in videos obtained by camera traps. Given the need for a large number of annotated images of the target animals to train the algorithms, we tested the incorporation of auxiliary data to improve detection models, fine-tuning the YOLOv10 framework using varying proportions of them. The improvement of these automatic detection techniques contributes to conservation efforts, by providing automatic tools to monitor solutions that minimize the impact of human interference in animals habitats.

arXiv:2607.01307v1 Announce Type: new Abstract: NA methylation profiling has become a powerful approach for central nervous system (CNS) tumor classification, yet important challenges remain regarding cross-cohort transferability, methodological correctness, and robust multiclass evaluation. In this work, we propose a novel and methodologically rigorous machine-learning approach for methylation-based CNS tumor classification that combines Sparse Random Projection for dimensionality reduction with multinomial logistic regression for classification. We evaluate the proposed approach in the same general experimental setting established by a widely used reference classifier. On the 2,801-sample reference cohort, our method achieves a mean accuracy of 96\% under stratified 3-fold cross-validation. On the independent 1,104-sample clinical evaluation cohort, it reaches 86\% accuracy at the 91-class level and 93\% when predictions are evaluated at the methylation class family level. These results improve upon the corresponding state-of-the-art reference figures of 82\% class-level concordance and 88\% family-level concordance, yielding absolute gains of approximately 4 and 5 percentage points, respectively. This improvement is clinically relevant: in a diagnostic setting, a 5-point increase in correct tumor classification can directly affect cancer subtype assignment and, in turn, influence treatment selection and downstream clinical decision-making. Our results show that the proposed model, grounded in stronger methodological practice in machine learning, consistently outperforms the previous state of the art across evaluation settings and can materially improve the reliability of CNS tumor classification.

arXiv:2607.01239v1 Announce Type: new Abstract: Character-level perturbations bypass safety alignment in modern LLMs despite leaving prompts human-readable. We identify and test a central structural mechanism: BPE tokenization fragments safety-critical words into sub-word pieces, and the three public alignment datasets we surveyed contain no intentionally fragmented inputs. The mechanism is a chain, tested end-to-end on five model families (Qwen-3-4B, Qwen-2.5-7B, Gemma-3-4B, Llama-3.1-8B, Mistral-7B). An optimization targeting safety-token fragmentation flips the first-token refusal trigger on 80-100% of refused HarmBench prompts, with 48% of those flips producing genuinely harmful outputs (per-model 29-65%; gap-vs-behavior ROC-AUC 0.66-0.98, pooled 0.84). Activation patching localizes the disrupted signal to the last ${\sim}30\%$ of layers; an alignment-data scan finds zero fragmented prompts among 30,000 examples (positive-control recall $\geq 99\%$ at attack-relevant intensities); and targeted-mutation experiments isolate safety words as the disruption locus. On the defense side, a 68-cell grid (55 trained checkpoints) shows that no DPO configuration achieves seed- and pool-stable ASR closure on the three families with closed pool-size confounds. SFT trained on fragmented prompts closes ASR on 3/5 families but only via global collapse that raises refusal on benign prompts as well, indicating the missing distribution is necessary but not sufficient under the LoRA-16 recipe we tested. To distinguish selective repair from global collapse, we introduce Conv-Benign, a candidate paired diagnostic. All ASR claims are 3-judge-calibrated (cell rankings stable across judges; absolute levels $\pm$18pp; see App.~B.13).

arXiv:2607.01279v1 Announce Type: new Abstract: Cross-subject EEG stress detection remains challenging because discriminative stress-related patterns are both subject-dependent and frequency-specific. Conventional Riemannian methods model spatial covariance mainly in the time domain, overlooking neural oscillations that are critical for high-level cognitive state decoding, while standard temporal tokenization often fragments inter-slice temporal coherence. To address these limitations, we propose \method{}, an Intra-Inter Riemannian Manifold Attention Network for EEG-based stress detection. \method{} constructs spatial covariance matrices independently at each frequency point and maps them to the SPD tangent space, preserving channel-wise geometry together with frequency-specific discriminative cues. It further introduces frequency cluster aggregation to select informative spectral components and reduce redundancy by forming compact, data-driven frequency clusters aligned with EEG rhythms. Finally, an intra-inter slice attention module adaptively integrates local slice-level spectral dynamics and global temporal context across EEG sequences. Experiments on three datasets show that \method{} consistently outperforms five state-of-the-art baselines, achieving up to 82.78\% balanced accuracy while remaining efficient with only 1.60M parameters and 31.95M FLOPs.

arXiv:2607.01238v1 Announce Type: new Abstract: Recent advances in speech synthesis have shifted from phoneme representations to direct grapheme modeling. While phonemes address the one-to-many mapping between text and acoustics, they rely on grapheme-to-phoneme (G2P) systems that fail to capture speaker-specific acoustic variation. Prior work demonstrates that grapheme-based models outperform phoneme-based systems at scale, but not in low-resource settings. In this paper, we propose SPARCLE, a speaker-aware grapheme representation model that enriches characters with their precise acoustic realizations. SPARCLE is trained with a contrastive objective to align graphemes with corresponding Wav2Vec2 acoustic representations while conditioned on speaker identity. The resulting model serves as a replacement to G2P systems for downstream text-to-speech (TTS) tasks. We demonstrate that SPARCLE improves generation quality, reducing word error rates by half in extreme low-resource settings compared to standard grapheme-based models.

arXiv:2607.01280v1 Announce Type: new Abstract: Programming-by-example systems infer programs from a small set of input-output examples. Robust PBE work usually models wrong examples as samples from a stochastic noise process and then minimizes an expected or empirical loss. This paper studies a different failure mode: an adversary who sees the synthesizer and chooses the examples whose corruption most damages the returned program. We formalize fixed-set worst-case corruption for finite PBE version spaces, implement exact-within-bounded-pool and heuristic corruption searches for a string-transformation DSL, and introduce version-space partition aggregation (VPA), a defense that synthesizes on disjoint example groups and votes by semantic signatures. The central claim is deliberately bounded and partly negative: low-margin PBE tasks have an adversarial robustness dimension that random-typo and noisy-PBE evaluations miss, while semantic partition aggregation helps only when the clean semantics keep a partition vote margin, which often fails on realistic tasks. Evidence from curated/generated DSL tasks, accepted public SyGuS PBE_SLIA slices, SYNTRA Playgol v2, and noisy-PBE objective baselines supports that boundary. One curated edit flips all 8 spike tasks while 200-trial typo, DSL-pool, and distance-matched random controls succeed on 10.3%, 11.0%, and 16.7%; generated margin-1 rows flip under budget 1 yet VPA recovers them; on public SyGuS the vote margin is near one, so an adaptive attacker drives VPA accuracy to zero; accepted public SyGuS slices move across exact-within-pool budget boundaries; and Playgol shows positive paired-bootstrap gaps against typo and same-pool random controls on the 141 accepted rows. A small exact-output prompt harness over 20 controlled margin-1 tasks shows the same qualitative clean-to-attacked pattern across local and API models, while it is treated as a scope check, not a broad LLM benchmark.