A demon can be a lot of different things depending on its backstory, but the consensus seems to be that it's a malevolent entity serving or using an evil power. These creatures are often depicted with horns, a tail, claws, and fangs, but on the other hand, none of these features are mandatory and demons can appear in a variety of forms.

Whether youre playing on the original Nintendo Switch, the sleek handheld-only Switch Lite or the upgraded OLED model with its stunning display, youre in for a gaming experience thats versatile, immersive and downright fun. The beauty of the Switch lineup is that, despite having different models, all games are compatible across the board. So no matter which version you own, youll have access to the entire library of games without worrying about what works on which console. This flexibility makes the Switch lineup ideal for both casual gamers and hardcore fans who love having the ability to access every title, whether theyre at home on the big screen or gaming on the go.With such a vast and diverse game library, choosing the best games can feel a bit overwhelming. Nintendo has something for everyone from epic action adventures and thrilling multiplayer battles to relaxing life sims and nostalgic platformers that bring back memories of classic gaming days. Whether youre looking to dive into sprawling RPGs, conquer puzzles or enjoy a bit of friendly competition, the Switch has you covered. Best Nintendo Switch games for 2025 Check out our entire Best Games series including the best Nintendo Switch games, the best PS5 games, the best Xbox games, the best PC games and the best free games you can play today. This article originally appeared on Engadget at https://www.engadget.com/gaming/nintendo/best-nintendo-switch-games-160029843.html?src=rss

This brand identity hits all the right notes.

Three big themes emerged within logo design in 2024.

MacworldWe all know Apple products are about sleek design, seamless integration, and simplifying life. Thats why we love the PowerBanda band for Apple Watches that doesnt just look good but powers your watch on the go.Say goodbye to tangled charging cables and clunky docks. The PowerBand features a built-in MagSafe-compatible charger, giving you a reliable, hassle-free way to keep your Apple Watch juiced up wherever life takes you.Ingeniously crafted, the ultra-thin charger is seamlessly integrated into the durable nylon weave band, offering an all-in-one charging solution thats lightweight and incredibly discreet. Just flip out the built-in USB charging connector, plug it into any outlet, and let PowerBand do the rest. No battery packs to managejust sleek, reliable convenience that fits perfectly with your Apple ecosystem.The nylon weave ensures a breathable, soft fit thats perfect for all-day wear, while the elastic pull tabs make adjusting the band quick.If youre looking for an easy way to stay charged on the go, this could be your go-to.For a limited time, the PowerBand is on sale for just $29.97 (reg. $49).PowerBand: Apple Watch Band with Built-In MagSafe Charger $29.97Get It HereStackSocial prices subject to change.

Code generation using Large Language Models (LLMs) has emerged as a critical research area, but generating accurate code for complex problems in a single attempt remains a significant challenge. Even skilled human developers often require multiple iterations of trial-and-error debugging to solve difficult programming problems. While LLMs have demonstrated impressive code generation capabilities, their self-debugging ability to analyze incorrect code and make necessary corrections is still limited. This limitation is evident in open-source models like StarCoder and CodeLlama, which show significantly lower self-refinement performance compared to models like GPT-3.5-Turbo.Existing approaches to improve code generation and debugging capabilities in LLMs have followed several distinct paths. LLMs have shown significant success across various code-related tasks, including code generation, bug fixing, program testing, and fuzzing. These models use extensive pre-training on vast datasets to understand patterns and generate contextually relevant code. However, most existing work has primarily focused on single-round generation rather than iterative improvement. Other methods like ILF, CYCLE, and Self-Edit have explored supervised fine-tuning approaches while solutions like OpenCodeInterpreter and EURUS have attempted to create high-quality multi-turn interaction datasets using advanced models for fine-tuning purposes.Researchers from Purdue University, AWS AI Labs, and the University of Virginia have proposed LEDEX (learning to self-debug and explain code), a novel training framework designed to enhance LLMs self-debugging capabilities. The framework builds on the observation that a sequential process of explaining incorrect code followed by refinement enables LLMs to analyze and improve faulty code in a better way. LEDEX implements an automated pipeline to collect high-quality datasets for code explanation, and refinement. Moreover, it combines supervised fine-tuning (SFT) and reinforcement learning (RL) approaches, utilizing successful and failed trajectories with a specialized reward system that evaluates code explanation and refinement quality.LEDEX employs a comprehensive architecture containing data collection, verification, and multi-stage training processes. The framework begins by collecting code explanation and refinement datasets through queries to pre-trained or instruction-tuned models. These responses undergo rigorous execution-based verification to filter and maintain only high-quality explanation and refinement data. The collected dataset then serves as input for supervised fine-tuning which significantly enhances the models capabilities in bug explanation and code refinement. LEDEX uses programming problems from MBPP, APPS, and CodeContests to train data. To expand the dataset of incorrect solutions, the framework prompts pre-trained LLMs like StarCoder and CodeLlama with 3-shot examples to generate 20 solutions per problem.LEDEX is evaluated using three model backbones: StarCoder-15B, CodeLlama-7B, and CodeLlama-13B, with initial training data collected from GPT-3.5-Turbo. The SFT phase shows significant improvements, achieving up to a 15.92% increase in pass@1 and 9.30% in pass@10 metrics across four benchmark datasets. The subsequent RL phase further enhances performance with additional improvements of up to 3.54% in pass@1 and 2.55% in pass@10. Notably, LEDEXs model-agnostic nature is shown through experiments with CodeLlama-7B, which achieve substantial improvements (8.25% in pass@1 and 2.14% in pass@10) even when trained on data collected from CodeLlama-34B or itself, proving its effectiveness independent of GPT-3.5-Turbo.In conclusion, researchers introduced LEDEX, a comprehensive and scalable framework that combines automated data collection, verification processes, SFT, and RL with innovative reward designs to significantly improve LLMs ability to identify and correct code errors. The frameworks model-agnostic nature is evidenced by its successful implementation with GPT-3.5-Turbo and CodeLlama, while its rigorous data verification process ensures the quality of code explanations and refinements. Human evaluations further validate the frameworks effectiveness, confirming that LEDEX-trained models produce superior code explanations that effectively assist developers in understanding and resolving code issues.Check out the Paper. All credit for this research goes to the researchers of this project. Also,dont forget to follow us onTwitter and join ourTelegram Channel andLinkedIn Group. Dont Forget to join our60k+ ML SubReddit. Sajjad Ansari+ postsSajjad Ansari is a final year undergraduate from IIT Kharagpur. As a Tech enthusiast, he delves into the practical applications of AI with a focus on understanding the impact of AI technologies and their real-world implications. He aims to articulate complex AI concepts in a clear and accessible manner. [Download] Evaluation of Large Language Model Vulnerabilities Report (Promoted)

This article contains spoilers for Squid Game season 2.Squid Game doesnt work without choice. The first season of Netflixs global hit made that clear with its standout second episode, Hell, in which contestants in the sprawling competition of deadly childrens games successfully vote to void their contract and exit the event.Without that single vote, Squid Game is little more than an exercise in gratuitous violence. A story in which rich VIPs kidnap members of Seouls desperate underclass and force them to do battle against one another plays a lot differently from a story that invites them in with the promise of a 45.6 billion payday.Or is it?Because, as the players in Squid Game season 1 discover much to their chagrin, sometimes a choice isnt much of a choice at all. The name of season 1 episode 2 Hell doesnt refer to the circumstances within the games but outside of them. Player 456 Seong Gi-hun (Lee Jung-jae) and his peers return to a real world so full of debts and threats that they all just opt to re-enter the games once again. In the end, they had the choice to leave and they had the choice to return. But that doesnt mean they felt in control of their destiny.That grappling with choice was the most thematically-rich aspect of Squid Games first season. Now, in its second season on Netflix, Squid Game returns to the topic once again in a far grander fashion. In the process, the show expands its trenchant critique of capitalism into a larger exploration of capital, the democratic process, and the increasingly exploitative ways they can intersect. Arriving at the tail end of an election year for much of the Western world, Squid Game once again feels like the timeliest show on television.The contestants of Squid Game season 2 arent afforded one choice, but several. At the conclusion of every game, the surviving players are invited to vote on whether they want the games to continue. Though he should know better from his voting experience in the 2021 version of the games, returning player 456 Gi-hun initially views this vote as a chance to save all of these wretched souls.Thanks to Gi-huns pre-existing knowledge of the first challenge, Red Light, Green Light, the squid gamers dont suffer too many casualties. With only 91 players eliminated, 365 remain to play another day. To Gi-hun, 365 souls saved is a massive victory. To those 365, however, theres another, more important number: 9.1 billion. Thats the amount of money available in prize winnings after the 91 deaths in the first challenge. If the remaining contestants exit the challenge now, they will take home only 24 million won or roughly $16,000 per person.Additionally, Gi-huns mere presence as a survivor of the games only confirms to the players that winning them is a possibility. And doing so will be a lot easier with him around. We have a previous winner here. What do we have to worry about? Player 100 (who is 10 billion in debt) says. Once again, the designers of the games have revealed themselves to be keen understanders of the human condition. Not having choice involved wouldnt be as satisfying to the contests spectators even the hyper rich dont want to feel like needless butchers. But the element of choice has to be present in such as way as to not actually jeopardize the continuation of the games. In this way, Gi-huns involvement in another version of the games feels less like his attempt to end them and more like the Front Man using him as a carrot to dangle to keep them going.Through the three rounds of voting featured in season 2, Squid Game takes on the rhythms of a functioning and healthy democracy. Voters are given X and O badges depending on which vote they cast and they quickly split into factions that resemble familiar two-party political structures. The Xs and Os dine together, develop strategies, and even work on catchphrases with One more game! becoming a rallying cry of the Os. But regardless of the votes cast and any progress made, each election always inevitably ends in another round of slaughter.Its easy for the viewer to graft their own democratic frustrations onto the proceedings, particularly during what is election year for many in the West. While undoubtedly the most equitable political system weve come across yet, democracy can be a profoundly frustrating beast. Sartre had it right that hell is other people (hat tip to Squid Game season 1 episode 2 once again) and nothing brings you into closer contact with other people quite like the democratic process. Watching Gi-hun grapple with a barracks full of the ultimate low information voters can be downright triggering. What do you mean you want to play one more game with a 50% mortality rate? And how do you not know what a tariff is???Join our mailing listGet the best of Den of Geek delivered right to your inbox!For the poorly informed (i.e. most people), voting amounts to answering the same question over and over again: Are you upset? Y/N. A no vote maintains the status quo, while a yes vote hands the reins of power over to something that could be marginally betteror far worse. Voters in the Western world were particularly upset in the post-Covid era. Incumbency was a massive disadvantage in virtually every election from 2021 through 2024, with existing executive administrations losing elections in Poland, Argentina, South Africa, Mexico, the United Kingdom, the United States, and yes, South Korea.Normally, I would feel embarrassed to bring my American bias into an article about a South Korean series. But Squid Game creator Hwang Dong-hyuk, bless him, seems just as preoccupied with the United States in season 2 as anything else. One of the characters here, Player 230 (Choi Su-bong), even goes by the stage name Thanos, and peppers his language full of American English phrases. American culture looms large in the West and so too do American anxieties. One anxiety in particular, helps explain both the 2024 U.S. election cycle and Squid Game season 2: the fear of missing out.Comparatively to the rest of the globe, the post-Covid economic malaise didnt hit the United States too hard. Though inflation did spike, it quickly rebounded into shape while unemployment remained at a steady low. And yet, American voters still reacted similarly to their global peers by striving for something different, in this case, by returning to something familiar. So strong is the pull of change that one of the worlds most familiar names and faces can feel new again. We survived last time, so surely one more game cant hurt. That brings us back to capital once again. Squid Game season 1s satire of unrestrained capitalism came through loud and clear and met with little critical resistance. Releasing just months after supply chains fell during the pandemic, Squid Game found a receptive audience to the concept that maybe the free markets were a little too free. Now season 2 gets even bolder and more challenging by tying capital directly into its more popular cousin, democracy.No one involved is claiming that democracy is bad or less preferable to an autocratic system. But Squid Game merely notes that democracy does lend itself to unregulated capitalism in a way that capitalists would prefer you didnt notice. Some estimates claim that approximately $15.9 billion was spent on the 2024 U.S. presidential elections. The country (and worlds) richest man, Elon Musk, spent $277 million alone on the effort to elect President Trump. You could even bump that number up by $44 billion if you consider Musks purchasing of Twitter to create a right-wing propaganda apparatus as a campaign contribution.That is undoubtedly a lot of money, donated freely and legally thanks to the Supreme Courts (dogshit) ruling in 2010 court case Citizens United v. Federal Election Commission. But its also likenot that much money in the grand scheme of things. And when it comes to influencing the most powerful executive office in the world, its a downright bargain for any enterprising oligarch with some disposable cash to spend and some capital gains taxes to slash. In fact, its probably not too far off from the money spent each time on prize winnings and operations of the squid games. One thing you cant say about Squid Games VIPs is that they skimp on production value.The nice thing about both Squid Game season 2s depiction of democracy and our own is that neither of them have ended yet. Gi-hun and his allies wisely come to realize that sham elections are getting them nowhere and decide to take matters into their own hands by staging an armed revolution. In the real life version of Squid Games own South Korea, citizens took to the streets in early December to defend their democracy against President Yoon Suk Yeols unlawful declaration of martial law.In the end, the problem is never only capitalism or democracy. Its people. Thankfully, they can be the solution too.All seven episodes of Squid Game season 2 are available to stream on Netflix now.

Dec 27, 2024Ravie LakshmananFirewall Security / VulnerabilityPalo Alto Networks has disclosed a high-severity vulnerability impacting PAN-OS software that could cause a denial-of-service (DoS) condition on susceptible devices.The flaw, tracked as CVE-2024-3393 (CVSS score: 8.7), impacts PAN-OS versions 10.X and 11.X, as well as Prisma Access running PAN-OS versions. It has been addressed in PAN-OS 10.1.14-h8, PAN-OS 10.2.10-h12, PAN-OS 11.1.5, PAN-OS 11.2.3, and all later PAN-OS versions."A denial-of-service vulnerability in the DNS Security feature of Palo Alto Networks PAN-OS software allows an unauthenticated attacker to send a malicious packet through the data plane of the firewall that reboots the firewall," the company said in a Friday advisory."Repeated attempts to trigger this condition will cause the firewall to enter maintenance mode."Palo Alto Networks said it discovered the flaw in production use, and that it's aware of customers "experiencing this denial-of-service (DoS) when their firewall blocks malicious DNS packets that trigger this issue."The extent of the activity is presently unknown. The Hacker News has reached out to Palo Alto Networks for further comment, and we will update the story if we hear back.It's worth pointing out that firewalls that have the DNS Security logging enabled are affected by CVE-2024-3393. The severity of the flaw also drops to a CVSS score of 7.1 when access is only provided to authenticated end users via Prisma Access.The fixes have also been extended to other commonly deployed maintenance releases -PAN-OS 11.1 (11.1.2-h16, 11.1.3-h13, 11.1.4-h7, and 11.1.5)PAN-OS 10.2 (10.2.8-h19, 10.2.9-h19, 10.2.10-h12, 10.2.11-h10, 10.2.12-h4, 10.2.13-h2, and 10.2.14)PAN-OS 10.1 (10.1.14-h8 and 10.1.15)PAN-OS 10.2.9-h19 and 10.2.10-h12 (only applicable to Prisma Access)PAN-OS 11.0 (No fix owing to it reaching end-of-life status on November 17, 2024)As workarounds and mitigations for unmanaged firewalls or those managed by Panorama, customers have the option of setting Log Severity to "none" for all configured DNS Security categories for each Anti-Spyware profile by navigating to Objects > Security Profiles > Anti-spyware > (select a profile) > DNS Policies > DNS Security.For firewalls managed by Strata Cloud Manager (SCM), users can either follow the above steps to disable DNS Security logging directly on each device, or across all of them by opening a support case. For Prisma Access tenants managed by SCM, it's recommended to open a support case to turn off logging until an upgrade is carried out.Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.SHARE

Dec 27, 2024Ravie LakshmananBotnet / DDoS AttackCybersecurity researchers are warning about a spike in malicious activity that involves roping vulnerable D-Link routers into two different botnets, a Mirai variant dubbed FICORA and a Kaiten (aka Tsunami) variant called CAPSAICIN."These botnets are frequently spread through documented D-Link vulnerabilities that allow remote attackers to execute malicious commands via a GetDeviceSettings action on the HNAP (Home Network Administration Protocol) interface," Fortinet FortiGuard Labs researcher Vincent Li said in a Thursday analysis."This HNAP weakness was first exposed almost a decade ago, with numerous devices affected by a variety of CVE numbers, including CVE-2015-2051, CVE-2019-10891, CVE-2022-37056, and CVE-2024-33112."According to the cybersecurity company's telemetry data, attacks involving FICORA have targeted various countries globally, whereas those related to CAPSAICIN primarily singled out East Asian territories like Japan and Taiwan. The CAPSAICIN activity is also said to have been "intensely" active only between October 21 and 22, 2024.FICORA botnet attacks lead to the deployment of a downloader shell script ("multi") from a remote server ("103.149.87[.]69"), which then proceeds to download the main payload for different Linux architectures separately using wget, ftpget, curl, and tftp commands.Present within the botnet malware is a brute-force attack function containing a hard-coded list of usernames and passwords. The Mirai derivative also packs in features to conduct distributed denial-of-service (DDoS) attacks using UDP, TCP, and DNS protocols.The downloader script ("bins.sh") for CAPSAICIN leverages a different IP address ("87.10.220[.]221"), and follows the same approach to fetch the botnet for various Linux architectures to ensure maximum compatibility."The malware kills known botnet processes to ensure it is the only botnet executing on the victim host," Li said. "'CAPSAICIN' establishes a connection socket with its C2 server, '192.110.247[.]46,' and sends the victim host's OS information and the nickname given by the malware back to the C2 server."CAPSAICIN then awaits for further commands to be executed on the compromised devices, including "PRIVMSG," a command that could be used to perform various malicious operations such as follows -GETIP - Get the IP address from an interfaceCLEARHISTORY - Remove command historyFASTFLUX - Start a proxy to a port on another IP to an interfaceRNDNICK - Randomize the victim hosts' nicknameNICK - Change the nickname of the victim hostSERVER - Change command-and-control serverENABLE - Enable the botKILL - Kill the sessionGET - Download a fileVERSION - Requests version of the victim hostIRC - Forward a message to the serverSH - Execute shell commandsISH - Interact with victim host's shellSHD - Execute shell command and ignore signalsINSTALL - Download and install a binary to "/var/bin"BASH - Execute commands using bashBINUPDATE - Update a binary to "/var/bin" via getLOCKUP - Kill Telnet backdoor and execute the malware insteadHELP - Display help information about the malwareSTD - Flooding attack with random hard-coded strings for the port number and target specified by the attackerUNKNOWN - UDP flooding attack with random characters for the port number and target specified by the attackerHTTP - HTTP flooding attack.HOLD - TCP connection flooding attack.JUNK - TCP flooding attack.BLACKNURSE - BlackNurse attack, which is based on the ICMP packet flooding attackDNS - DNS amplification flooding attackKILLALL - Stop all DDoS attacksKILLMYEYEPEEUSINGHOIC - Terminate the original malware"Although the weaknesses exploited in this attack had been exposed and patched nearly a decade ago, these attacks have remained continuously active worldwide," Li said. "It is crucial for every enterprise to regularly update the kernel of their devices and maintain comprehensive monitoring."Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.SHARE