Tootfinder

Defending the Black Sea: New Strategies to Counter Russian Threats: https://benborges.xyz/2024/03/05/defending-the-black.html

people reading orwell: it's called the ministry of TRUTH when in reality all it does is spreading LIES!!!
those same people: it's the ministry of DEFENSE so it must be DEFENDING and why shouldn't they have a right to DEFEND themselves?

AFC West Division odds: Defending champion Chiefs open as favorites https://www.foxsports.com/stories/nfl/afc-west-division-odds-chiefs-raiders-chargers-broncos

AFC West Division odds: Defending champion Chiefs open as favorites
Kansas City is favored to win the AFC West next season, which would equal its ninth straight division title. Check out the early odds.

Besides the fact that their lips are moving?
Here's How We Know Republicans Are Lying About Their 'Support' For #IVF - National Memo
https://www.nationalmemo.com/ivf-republica

Here's How We Know Republicans Are Lying About Their 'Support' For IVF
Rep. Nancy Mace has taken the bold step of defending in vitro fertilization by introducing a nonbinding resolution. That way Republicans in the House can pretend that they’re taking steps to protect IVF without actually protecting IVF.

Doubling down on dogicide is definitely a political strategy for these weary times.
From: @…
https://press.coop/@nytimes/112389566497954276…

The New York Times :press: (@nytimes@press.coop)
Attached: 1 image Kristi Noem Suggests Biden’s Dog Should Have Been Killed, Too The South Dakota governor, defending her tale of shooting and killing her family’s dog, suggested that President Biden’s German shepherd, Commander, had merited a similar fate. #press https://www.nytimes.com/2024/05/05/us/politics/kristi-noem-biden-dog.html?utm_source=press.coop

Dog caught the car. Doesn’t know what to do! They’re trapped. Senators Now Defending IVF CoSponsored Bill to Outlaw It - TPM – Talking Points Memo https://talkingpointsmemo.com/edblog/senators-now-defending-ivf-cosponsored-bill-to-outlaw-it<…

Senators Now Defending IVF CoSponsored Bill to Outlaw It
Sen. Steve Daines (R) of Montana heads the Senate Republican campaign committee, which today sent out a memo to all its candidates instructing them to vigorously defend IVF fertility treatments which…

From Attack to Defense: Insights into Deep Learning Security Measures in Black-Box Settings
Firuz Juraev, Mohammed Abuhamad, Eric Chan-Tin, George K. Thiruvathukal, Tamer Abuhmed
https://arxiv.org/abs/2405.01963

From Attack to Defense: Insights into Deep Learning Security Measures in Black-Box Settings
Deep Learning (DL) is rapidly maturing to the point that it can be used in safety- and security-crucial applications. However, adversarial samples, which are undetectable to the human eye, pose a serious threat that can cause the model to misbehave and compromise the performance of such applications. Addressing the robustness of DL models has become crucial to understanding and defending against adversarial attacks. In this study, we perform comprehensive experiments to examine the effect of ad…

BBC is interviewing an ex-Clinton aide who said that some of the people are “Anarchists”. The interviewer challenges "are you seriously saying these are anarchists?”....
“I didn't say all…" says guy. Then he spends the rest of the time referring to them as anarchists. This is classic.
Guy is just going off. He says: “There is no such thing as genocide from Israel because Hamas has a policy of genocide... apartheid is not a thing…
"These anarchists at Co…

Doubling down on dogicide is definitely a political strategy for these weary times.
From: @…
https://press.coop/@nytimes/112389566497954276…

The New York Times :press: (@nytimes@press.coop)
Attached: 1 image Kristi Noem Suggests Biden’s Dog Should Have Been Killed, Too The South Dakota governor, defending her tale of shooting and killing her family’s dog, suggested that President Biden’s German shepherd, Commander, had merited a similar fate. #press https://www.nytimes.com/2024/05/05/us/politics/kristi-noem-biden-dog.html?utm_source=press.coop

2022 retro-link! https://blog.apnic.net/2022/07/22/why-the-nro-is-defending-afrinic/ - Why the NRO is defending AFRINIC.

Court orders Trader Joe's to pay workers' union $112,622.12 in attorneys' fees for defending against the "exceptionally weak" trademark lawsuit:
"Employers should be discouraged from bringing meritless claims against unions they are challenging at the ballot box."

This https://arxiv.org/abs/2402.16459 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Defending LLMs against Jailbreaking Attacks via Backtranslation
Although many large language models (LLMs) have been trained to refuse harmful requests, they are still vulnerable to jailbreaking attacks, which rewrite the original prompt to conceal its harmful intent. In this paper, we propose a new method for defending LLMs against jailbreaking attacks by ``backtranslation''. Specifically, given an initial response generated by the target LLM from an input prompt, our backtranslation prompts a language model to infer an input prompt that can lead to the re…

Law Firm Defending Trump Seeks to Withdraw From a Long-Running Case (Ken Bensinger/New York Times)
https://www.nytimes.com/2024/04/30/us/politics/trump-lawyers-delgado.html
http://www.memeorandum.com/240430/p88#a240430p88

Defending-champion Dalvin Cook, 13 other NFL stars participating in 'The Catch' competition on Saturday https://www.nfl.com/news/defending-champion-dalvin-cook-13-other-nfl-stars-participating-in-the-c…

Defending-champion Dalvin Cook, 13 other NFL stars participating in 'The Catch' competition on Saturday
Along with 13 other NFL stars, last year's winner Dalvin Cook is slated to participate in the Sport Fishing Championship's "The Catch" tournament, which airs Saturday, April 20 on CBS Sports Network.

Here's my latest Human Meme podcast episode: Defending Helen Keller: Debunking Myths and Championing Disability Rights - http://sites.libsyn.com/84966/defending-helen-keller-debunking-myths-and-championing-disability-rights

Stephen A. Smith injured his ankle defending Micah Parsons at All-Star Game https://www.yardbarker.com/nba/articles/stephen_a_smith_injured_his_ankle_defending_micah_parsons_at_all_star_game/s1_127_399748…

Stephen A. Smith injured his ankle defending Micah Parsons at All-Star Game
Stephen A. Smith is coaching one of the teams in Friday night’s NBA Celebrity All-Star Game, and the longtime ESPN analyst will be doing so in an ankle brace. At the start of Friday morning’s edition of “First Take,” host Molly Qerim revealed that she was told by Smith that Stephen A. suffered an ankle injury while practicing against Dallas Cowboys star Micah Parsons — who is on Smith’s team — on Thursday. “Micah crossed him over, and all I know is that Stephen A. is at the hosp…

Not just a story of a badass #plankton researcher, but also some interesting background about the #Ukraine Vernadsky #Antarctic research base and its link to the British Antarctic Survey. Two jobs, studyin…

Republicans Seem Completely OK With Trump’s Alarming Russia Comments
They are defending Donald Trump’s terrifying remarks that Russia can do “whatever the hell it wants.”
Conservatives once again fell in line behind Donald Trump on Monday, signing off on comments the GOP front-runner made over the weekend that read more like veiled threats against nations he deems are failing NATO’s guidelines.
“No, I would not protect you,” Trump told a crowd in Conway, North Carolina, on …

Republicans Seem Completely OK With Trump’s Alarming Russia Comments
Republicans are defending Donald Trump’s terrifying remarks that Russia can do “whatever the hell it wants.”

'His conduct gets much worse': Impeachment lawyer has a warning about Trump - Raw Story
https://www.rawstory.com/trump-2024-2667410801ets-worse/

'His conduct gets much worse': Impeachment lawyer has a warning about Trump
Former chief impeachment lawyer Barry Berke brought up the former staff of Donald Trump on Friday, taking to the media to sound the alarm that a second term would be dangerous. The conversation began about the Trump trials and the notion that Americans deserve to see the evidence against him before ...

Just a horrid stretch of defending and I take it back about Quansah. That was extremely poor. #LFC

Russia sentences Meta spokesperson Andy Stone to six years in absentia for "publicly defending terrorism"; Russia designates Meta as an extremist organization (Reuters)
https://www.reuters.com/technology/russian…

Today the ambulance has been delivered to the front line: https://benborges.xyz/2024/02/29/today-the-ambulance.html

"If you were in charge of defending Trump...." says Rachel Maddow?
🤔 I'd be a donkey in elephant's clothing.

"Too often these days I find myself in the position of defending someone I think is annoying from someone I know is dangerous." The Polycrisis: Why can’t we stop talking about nonmonogamy? - The Yale Review https://yalereview.org/article/brandy-jensen-polycrisis

This https://arxiv.org/abs/2405.00289 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Adversarial Attacks and Defense for Conversation Entailment Task
Large language models (LLMs) that are proved to be very powerful on different NLP tasks. However, there are still many ways to attack the model with very low costs. How to defend the model becomes an important problem. In our work, we treat adversarial attack results as a new (unseen) domain of the model, and we frame the defending problem into how to improve the robustness of the model on the new domain. We focus on the task of conversation entailment, where multi-turn natural language dialogu…

This https://arxiv.org/abs/2403.05030 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Defending Against Unforeseen Failure Modes with Latent Adversarial Training
AI systems sometimes exhibit harmful unintended behaviors post-deployment. This is often despite extensive diagnostics and debugging by developers. Minimizing risks from models is challenging because the attack surface is so large. It is not tractable to exhaustively search for inputs that may cause a model to fail. Red-teaming and adversarial training (AT) are commonly used to make AI systems more robust. However, they have not been sufficient to avoid many real-world failure modes that differ…

‘Extreme’ US anti-abortion group ramps up lobbying in Westminster
The UK branch of the Alliance Defending Freedom has increased its spending and is forging ties with key [mostly Tory] MPs
https://www.theguardian.com/world/2024/apr

We don't have a lot of domestic rituals around here but one we do have is every year about now, or really usually earlier - it's been wet - I notice that the lawn is languishing and it must be time to turn on the sprinklers for the season. Or rather, it's time to figure out in what crazy ways the sprinkler system has become busted since - *checks watch* - late October. Plants grow in front of the heads, water stops coming to the heads, or the heads just spontaneously disintegrate from sunlight losing pressure in the whole circuit. They sell slightly more robust partly-brass heads but those bust too, and you're never going to be able to fix one of those by massaging it like you can half the time with the plastic heads. This year, amazingly, 90% of everything is 90% working which is good enough. The odds of the sprinkler computer surviving through the Winter un-reset are low, and they didn't come through this time around, so I had to reset everything. I am covered with little cuts from the six types of unidentifiable spiky vegetation I'm sure we never planted that was defending sprinkler heads it had seized, to keep all the water and leave none for our unassuming monsteras.

Hah! "...clearly haven't got the message that the new Government is wanton..." https://www.rnz.co.nz/news/national/509317/niwa-spends-700k-on-four-luxury-petrol-utes (I suspect that's a malapropism)

The Anti-Gay, Anti-Trans and Anti-Abortion Group Behind Those ‘He Gets Us’ Super Bowl Commercials - Ms. Magazine
https://msmagazine.com/2024/02/15/he-gets-us-hobby-lobby-trans-abortion-women-alliance-defending-freedom/

The Anti-Gay, Anti-Trans and Anti-Abortion Group Behind Those ‘He Gets Us’ Super Bowl Commercials - Ms. Magazine
The Anti-Gay, Anti-Trans and Anti-Abortion Group Behind Those ‘He Gets Us’ Super Bowl Commercials is Alliance Defending Freedom (ADF).

Social revolution, a how-to by @… (i know, it's NOT a how-to, fool), i loved reading this and was happy the campfire metaphor was not taken to a place i feared, the pissing on the embers before moving to the next campground:
"Social movements have an uphill battle. You’ve survived the initial repression, then you’ve survived the constant one-two punch of repression and recuperation. You’ve survived splintering by refusing to let the more radical elements within your movement be isolated and picked off. You’ve become a full-fledged revolution, combining mass social movements with militant actors who are redistributing property and defending the gains of the revolution.
The final recuperative force generally comes from within the revolutionary movement itself: authoritarianism. It’s Boromir grabbing the ring of power, it’s the Bolsheviks declaring themselves in charge of a pluralistic movement and killing everyone who disagrees with them.
The purpose of a social revolution is to redistribute power horizontally, but there will always be people who say “yes, and I should be given all of the power so that I can better redistribute it,” rather than continuing to work from the bottom-up model that built the revolution in the first place."
#revolution #anarchism

Hey! Just logged in (on the computer, after being around on the phone) and this, by @…, is the post at the very top of the feed - https://journa.host/@ScottLucas/112145

Which game should kick off the 2024 NFL season? Ravens-Chiefs among top options https://www.nfl.com/news/which-game-should-kick-off-the-2024-nfl-season-ravens-chiefs-among-top-options

Which game should kick off the 2024 NFL season? Ravens-Chiefs among top options
While we wait for the 2024 NFL schedule to be unveiled, NFL.com analysts debate which team they'd want to see face the defending Super Bowl champion Chiefs in the NFL Kickoff Game.

This https://arxiv.org/abs/2403.05030 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Defending Against Unforeseen Failure Modes with Latent Adversarial Training
AI systems sometimes exhibit harmful unintended behaviors post-deployment. This is often despite extensive diagnostics and debugging by developers. Minimizing risks from models is challenging because the attack surface is so large. It is not tractable to exhaustively search for inputs that may cause a model to fail. Red-teaming and adversarial training (AT) are commonly used to make AI systems more robust. However, they have not been sufficient to avoid many real-world failure modes that differ…

‘Honest and brave’: progressive Jewish figures defend Jonathan Glazer speech https://www.theguardian.com/film/2024/mar/20/jonathan-glazer-speech-jewish-groups-defense?CMP=Share_iOSApp_Other

Good for Paul Waldman and Thomas Schaller. They are defending their book. And they are right.
https://newrepublic.com/article/180570/trump-rural-white-resentment-honest-assessment

An Honest Assessment of Rural White Resentment Is Long Overdue
We say very clearly that rural America is hurting. But we refuse to justify attitudes that some scholars try to underplay.

classic case of doing the right thing for the wrong reason: https://news.slashdot.org/story/24/04/22/158257/russian-court-sentences-meta-spokesperson-to-six-years-in-absentia-calls-meta-…

Russian Court Sentences Meta Spokesperson To Six Years in Absentia, Calls Meta 'Extremist Organisation' - Slashdot
A military court in Moscow on Monday sentenced Meta spokesperson Andy Stone to six years in prison for "publicly defending terrorism," a verdict handed down in absentia, RIA news agency reported. Reuters: Meta itself is designated an extremist organisation in Russia and its Facebook and Instagram so...

Each of these is a whole field of study — there's a whole microcosm in metaphysics of language. But again and again, what truly matters beyond the effects and perception of our words and actions?
Many of us spend a lot of time caught in our heads thinking about words and labels and boxes and boundaries. I'm not suggesting that we don't need a bit of Wittgenstein's ladder to build our internal understanding of what we want, but at some point we have to get out of our own heads.
What truly matters beyond the effects and perception of our words and actions?
Does my gender matter beyond how I want to be treated? Does my sexuality matter beyond whether and who and how I want to have sex?
Do software delivery dates or OKRs matter beyond whether the users are happy or whether the software creates the effects we intend?
Does it matter to label a country's actions as "preventative" or "defensive" or "defending managed democracy" or "genocidal" (sorry Helldivers) when large numbers of living beings are dying?

While this is very problematic behavior one shouldn't ignore that the authors are all non-native speakers who might have used "AI" to sound more "white" and therefore "respectable". Because that's what "AI" text generators are: They make your text more white, more like the culturally dominant force. "AI" always has this very colonialist bend.
(Again, not defending the act just trying to add some potential context.)

Ionica Smeets (@ionica@mathstodon.xyz)
Attached: 1 image Subtle ways to spot AI might have written the introduction 😀 https://www.sciencedirect.com/science/article/abs/pii/S2468023024002402

This "US gov report bashing C and Stroustrup rebuttal defending C " is going to be an annual ritual, isn't it? 🥺
https://www.infoworld.com/article/3714401/c-plus-plus-creator-rebuts-white-house-warning.html<…

C++ creator rebuts White House warning
Biden administration seems oblivious of the strengths of contemporary C++ and the efforts to provide strong safety guarantees, Bjarne Stroustrup said.

Just a horrid stretch of defending and I take it back about Quansah. That was extremely poor. #LFC

"Methods of defending your local #library from government cuts" by #TomGauld: https://

Selenskyj fordert gleiche Einigkeit bei Unterstützung von Ukraine
Der ukrainische Präsident Wolodymyr Selenskyj fordert nach dem iranischen Angriff auf Israel seine Verbündeten auf, sein Land mit der gleichen Einigkeit zu unterstützen wie Israel. "Die ganze Welt war Zeuge des Einsatzes der Alliierten am Himmel über Israel und den Nachbarländern", so Selenskyj. Niemand sei in den Krieg hineingezogen …
🐦

Volodymyr Zelenskyy / Володимир Зеленський (@ZelenskyyUa) on X
The entire world witnessed allied action in the skies above Israel and neighboring countries. It demonstrated how truly effective unity in defending against terror can be when it is based on sufficient political will. Israel, the United States, the United Kingdom, France, and…

"If you were in charge of defending Trump...." says Rachel Maddow?
🤔 I'd be a donkey in elephant's clothing.

Scotland are defending brilliantly, but I'm not getting those 40 mins back. Bloody awful to watch. #irevsco #rugby #6nations

U.S. Pressure Helped Save Brazil’s Democracy
Biden’s stated commitment to defending democracy worldwide is often brushed off as mere rhetoric.
Yet, two weeks ago, Brazil’s federal police launched a high-profile raid against former President Jair Bolsonaro and more than 10 of his allies, including Brazil’s former navy chief, national security advisor, and ministers of defense and justice. Authorities accused the group of plotting a potential coup after Bolsonaro’s failed 2022 ree…

#russia declares Estonian PM Kaja Kallas a 'wanted' person. Are they preparing another “special military operation”?
https://www.dw.com/en/russia-declares-esto

Russia declares Estonian PM Kaja Kallas a 'wanted' person
The criminal charges reportedly relate to the destruction of Soviet-era monuments. Kaja Kallas has been a staunch supporter of Ukraine defending itself against the Russian invasion.

For all people not familiar with NATO treaty, there are no binding articles about how much members should spend on defense. Ofcourse they should spend more , no doubt about that. But the whole treaty is based on defending each other and Trump is actively undermining that principle, helping US main adversaries
#nato #trump

Ukraine Is Now 'Holy War,' Russian Church Declares
https://www.newsweek.com/russia-ukraine-war-holy-1884577

Ukraine Is Now 'Holy War,' Russian Church Declares
The Russian Orthodox Church said that the war in Ukraine is "defending the single spiritual space of Holy Russia."

This https://arxiv.org/abs/2402.16192 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing
Aligned large language models (LLMs) are vulnerable to jailbreaking attacks, which bypass the safeguards of targeted LLMs and fool them into generating objectionable content. While initial defenses show promise against token-based threat models, there do not exist defenses that provide robustness against semantic attacks and avoid unfavorable trade-offs between robustness and nominal performance. To meet this need, we propose SEMANTICSMOOTH, a smoothing-based defense that aggregates the predict…

🎾 @… #Tennis
Via José Morgado 🤖 @…
Draw in Buenos Aires, where Carlos Alcaraz is the defending champ!

Series B, Episode 06 - Trial
SAMOR: So recognized. You may stand at ease. [Travis does so.] Do you accept the authority of the Arbiter, the balance of the judgment program, and the competence of the defending officer?
TRAVIS: [Pause] I do.
https://blake.torpidity.net/s/206/46 📺 B7B3

May I again express my skepticism that what #Israel has been doing is just defending itself. https://www.nytimes.com/2024/04/26/world/middl…

Rabbis Arrested Near Gaza-Israel Border at Rally to Highlight Starvation
A group of about 30 rabbis and peace activists from Israel and the United States were trying to take food supplies into the territory.

Recapping Week 1 in the UFL: Stallions prevail in battle of champions https://www.espn.com/united-football-league/story/_/id/39839488/ufl-week-1-football-highlights-game-recaps-top-players

UFL Week 1: Highlights, game recaps, top players - ESPN
The spring football season started Saturday with the USFL champion Birmingham Stallions outlasting Arlington, the defending XFL champs.

Revealed: Mike Johnson's 'alarming' ties to slavery-defending Christian extremists - Raw Story
https://www.rawstory.com/mike-johnson-2667131761/

Revealed: Mike Johnson's 'alarming' ties to slavery-defending Christian extremists
The Daily Beast has published a new report revealing ties between House Speaker Mike Johnson (R-LA) and Christian fundamentalists who believe slavery is justified by the Bible.Specifically, the report notes that Johnson has openly praised Christian nationalist David Barton and has said that Barton's...

Blue Team Cats!
/
Perplexity.ai
Blue teamers are responsible for defending against attacks, both real and simulated by red teams. They maintain and improve the organization's internal network defenses to protect against cyber threats. Their focus is on detection, prevention, and response to security incidents. Blue teamers specialize in monitoring network activity, analyzing security alerts, managing firewalls and endpoint security.
/

Please, read the MSF blog post linked to in @… 's post. Wherever you are, whatever country, you need to let your elected officials know that THEY need to read it and address the issues #MSF raises.
Israel is not defending itself anymore; Israel is not going after the …

Please, read the MSF blog post linked to in @… 's post. Wherever you are, whatever country, you need to let your elected officials know that THEY need to read it and address the issues #MSF raises.
Israel is not defending itself anymore; Israel is not going after the …

This https://arxiv.org/abs/2402.12168 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning
Recently, various parameter-efficient fine-tuning (PEFT) strategies for application to language models have been proposed and successfully implemented. However, this raises the question of whether PEFT, which only updates a limited set of model parameters, constitutes security vulnerabilities when confronted with weight-poisoning backdoor attacks. In this study, we show that PEFT is more susceptible to weight-poisoning backdoor attacks compared to the full-parameter fine-tuning method, with pre…

This https://arxiv.org/abs/2312.07364 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCV_…

Collapse-Aware Triplet Decoupling for Adversarially Robust Image Retrieval
Adversarial training has achieved substantial performance in defending image retrieval against adversarial examples. However, existing studies in deep metric learning (DML) still suffer from two major limitations: weak adversary and model collapse. In this paper, we address these two limitations by proposing collapse-aware triplet decoupling (CA-TRIDE). Specifically, TRIDE yields a strong adversary by spatially decoupling the perturbation targets into the anchor and the other candidates. Furthe…

A colleague who’s in the online advertising space explained to me that while WhatsApp messages are indeed end to end encrypted, the social graph that links you to your connections (ie the people you message with) is not encrypted and not private. Facebook sells that info, but most people incorrectly think that because messages are encrypted that everything about your WhatsApp messaging is private. It is not. I’m not defending this, just explaining it.

A rightwing Christian lobby group that wants abortion to be banned has forged ties with an adviser to the British prime minister and is drawing up ­policy briefings for politicians.
The UK branch of the US-based Alliance Defending Freedom ( #ADF ) has more than doubled its spending since 2020 and been appointed a stakeholder in a parliamentary group on religious freedoms in a role that grants it direct a…

Selenskyj fordert gleiche Einigkeit bei Unterstützung von Ukraine
Der ukrainische Präsident Wolodymyr Selenskyj fordert nach dem iranischen Angriff auf Israel seine Verbündeten auf, sein Land mit der gleichen Einigkeit zu unterstützen wie Israel. "Die ganze Welt war Zeuge des Einsatzes der Alliierten am Himmel über Israel und den Nachbarländern", so Selenskyj. Niemand sei in den Krieg hineingezogen …
🐦

Volodymyr Zelenskyy / Володимир Зеленський (@ZelenskyyUa) on X
The entire world witnessed allied action in the skies above Israel and neighboring countries. It demonstrated how truly effective unity in defending against terror can be when it is based on sufficient political will. Israel, the United States, the United Kingdom, France, and…

This https://arxiv.org/abs/2311.03338 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Defending a Static Target Point with a Slow Defender
This paper studies a target-defense game played between a slow defender and a fast attacker. The attacker wins the game if it reaches the target while avoiding the defender's capture disk. The defender wins the game by preventing the attacker from reaching the target, which includes reaching the target and containing it in the capture disk. Depending on the initial condition, the attacker must circumnavigate the defender's capture disk, resulting in a constrained trajectory. This condition prod…

Structure-dependent fragmentation risk of rubble-pile asteroids on low-impacts
Chenyang Huang, Yang Yu, Peter R. King, Bin Cheng, Raphael Blumenfeld
https://arxiv.org/abs/2402.16723

Structure-dependent fragmentation risk of rubble-pile asteroids on low-impacts
A key strategy in defending against stray asteroids is deflection from a collision trajectory by a low-momentum impact. This is to avoid a potential rain of large hazardous fragments, which high-momentum blasts may generate. Using a proof-of-principle numerical model, we show that even low-momentum impacts on rubble pile asteroids (RPAs), which most asteroids are, may lead to their fracturing into large fragments because of their internal non-uniform chain-like stress distribution. Since stress…

Series B, Episode 06 - Trial
SAMOR: So recognized. You may stand at ease. [Travis does so.] Do you accept the authority of the Arbiter, the balance of the judgment program, and the competence of the defending officer?
TRAVIS: [Pause] I do.
https://blake.torpidity.net/s/206/46 📺 B7B3

Antonio Pierce: Raiders must treat Patrick Mahomes like Michael Jordan https://www.reviewjournal.com/sports/raiders/antonio-pierce-raiders-must-treat-patrick-mahomes-like-michael-jordan-3004619/

Antonio Pierce: Raiders must treat Patrick Mahomes like Michael Jordan
Raiders coach Antonio Pierce referenced the physical treatment Michael Jordan received in the 1980s and 1990s for defending Chiefs quarterback Patrick Mahomes.

May I again express my skepticism that what #Israel has been doing is just defending itself. https://www.nytimes.com/2024/04/26/world/middl…

Rabbis Arrested Near Gaza-Israel Border at Rally to Highlight Starvation
A group of about 30 rabbis and peace activists from Israel and the United States were trying to take food supplies into the territory.

Anti-abortion attorneys ascend federal government ranks with Christian right legal training
The conservative Christian law firm Alliance Defending Freedom’s expansive ties include  federal judges and most recently Speaker of the House
The nonprofit law firm that helped Mississippi solicitor general Scott Stewart argue and win the #Dobbs case
— the Alliance Defending Freedom
— has work…

Anti-abortion attorneys ascend federal government ranks with Christian right legal training • Oregon Capital Chronicle
The U.S. house speaker is among a number of people affiliated with a conservative group that's trying to capture the courts.

Defending Against Indirect Prompt Injection Attacks With Spotlighting
Keegan Hines, Gary Lopez, Matthew Hall, Federico Zarfati, Yonatan Zunger, Emre Kiciman
https://arxiv.org/abs/2403.14720

Defending Against Indirect Prompt Injection Attacks With Spotlighting
Large Language Models (LLMs), while powerful, are built and trained to process a single text input. In common applications, multiple inputs can be processed by concatenating them together into a single stream of text. However, the LLM is unable to distinguish which sections of prompt belong to various input sources. Indirect prompt injection attacks take advantage of this vulnerability by embedding adversarial instructions into untrusted data being processed alongside user commands. Often, the …

Cowboys rank last in cash spending as Stephen Jones defends ‘all-in’ motto https://www.yardbarker.com/nfl/articles/cowboys_rank_last_in_cash_spending_as_stephen_jones_defends_all_in_motto/s1_17236_4024861…

Cowboys rank last in cash spending as Stephen Jones defends ‘all-in’ motto
As the Dallas Cowboys head into the 2024 NFL Draft next week, executive vice president Stephen Jones found himself, once again, defending the front office’s so-called “all-in” strategy. The Cowboys currently rank last in the league in average cash spending at $189 million. That figure, in comparison to the Cleveland Browns at No. 1 with $326 million, leads fans to believe the team isn’t quite as “all-in” as Jerry Jones would have them believe. The spending, on top of the team allowi…

It wasn’t a pretty half by any means, but #LFC even half asleep have more quality than United and certainly don’t lack the mental fortitude to come from a goal down.
LFC were their own worst enemy with poor defending on the right side and wayward passes under the duress of a press that really wasn’t all that troublesome.
Would like to see Bradley in for Gomez in the second half.

Ukrainian serviceman receives 20-year Russian prison sentence: https://benborges.xyz/2024/03/07/ukrainian-serviceman-receives.html

Defending against Data Poisoning Attacks in Federated Learning via User Elimination
Nick Galanis
https://arxiv.org/abs/2404.12778 https://

Defending against Data Poisoning Attacks in Federated Learning via User Elimination
In the evolving landscape of Federated Learning (FL), a new type of attacks concerns the research community, namely Data Poisoning Attacks, which threaten the model integrity by maliciously altering training data. This paper introduces a novel defensive framework focused on the strategic elimination of adversarial users within a federated model. We detect those anomalies in the aggregation phase of the Federated Algorithm, by integrating metadata gathered by the local training instances with Di…

Selenskyj ersetzt Armeechef Saluschnyj
Der ukrainische Präsident Wolodymyr Selenskyj hat seinen Armeechef Walerij Saluschnyj abgesetzt. Das teilte Selenskyj in einer Videobotschaft mit. Zuvor wurde bereits tagelang über eine mögliche Entlassung des Generals berichtet. Abgelöst werden soll dieser nun vom bisherigen Kommandanten der Landstreitkräfte Oleksandr Syrskyj. "Die Zeit für eine solche Erneueru…
🐦

Volodymyr Zelenskyy / Володимир Зеленський (@ZelenskyyUa) on X
I met with General Valerii Zaluzhnyi. I thanked him for the two years of defending Ukraine. We discussed the renewal that the Armed Forces of Ukraine require. We also discussed who could be part of the renewed leadership of the Armed Forces of Ukraine. The time for such a renewal…

The Carrizo Comecrudo endured slavery by the Spanish, indoctrination by Catholic mission schools, intrusion by Big Oil, and fragmentation by the Mexican border and by Trump’s border wall.
Now, they face the desecration of their sacred lands by liquefied natural gas plants and the SpaceX launch center, “Starbase.”
Last May, the tribe joined the environmental groups Save RGV and the Sierra Club in a lawsuit against Cameron County and the Texas General Land Office, stating that bloc…

Defending Native Sacred Sites From Elon Musk and SpaceX
The Carrizo Comecrudo Tribe of Texas is suing to protect the tribe’s ancestral land.

Enhancing IoT Security: A Novel Feature Engineering Approach for ML-Based Intrusion Detection Systems
Afsaneh Mahanipour, Hana Khamfroush
https://arxiv.org/abs/2404.19114 https://arxiv.org/pdf/2404.19114
arXiv:2404.19114v1 Announce Type: new
Abstract: The integration of Internet of Things (IoT) applications in our daily lives has led to a surge in data traffic, posing significant security challenges. IoT applications using cloud and edge computing are at higher risk of cyberattacks because of the expanded attack surface from distributed edge and cloud services, the vulnerability of IoT devices, and challenges in managing security across interconnected systems leading to oversights. This led to the rise of ML-based solutions for intrusion detection systems (IDSs), which have proven effective in enhancing network security and defending against diverse threats. However, ML-based IDS in IoT systems encounters challenges, particularly from noisy, redundant, and irrelevant features in varied IoT datasets, potentially impacting its performance. Therefore, reducing such features becomes crucial to enhance system performance and minimize computational costs. This paper focuses on improving the effectiveness of ML-based IDS at the edge level by introducing a novel method to find a balanced trade-off between cost and accuracy through the creation of informative features in a two-tier edge-user IoT environment. A hybrid Binary Quantum-inspired Artificial Bee Colony and Genetic Programming algorithm is utilized for this purpose. Three IoT intrusion detection datasets, namely NSL-KDD, UNSW-NB15, and BoT-IoT, are used for the evaluation of the proposed approach.

Cowboys’ Micah Parsons reiterates NFL’s incompetency with holding calls following Super Bowl havoc https://www.yardbarker.com/nfl/articles/cowboys_micah_parsons_reiterates_nfls_incom…

Cowboys’ Micah Parsons reiterates NFL’s incompetency with holding calls following Super Bowl havoc
The Super Bowl ended with the Kansas City Chiefs becoming one of the only two wins in NFL history to win back-to-back Vince Lombardi Trophy. In an exciting matchup against the San Francisco 49ers, the Chiefs won the game against all odds at 25-22 in overtime.  However, since the game, there have been quite a handful of theories that contradict the defending Super Bowl champions’ credibility towards the grand victory. Joining the queue is the NFL star defensive player, Micah Parsons, who does…

Republicans sidestep criticizing Trump’s attack on NATO
As they voted to advance a funding package for Ukraine and Israel on Sunday, some Republican senators pretended that they had not heard the president’s remarks.

“I haven’t seen that, so I’d need to listen to his quote first,” said Sen. Tommy Tuberville (R-Ala.). After a reporter asked him if he thought it was correct for Trump to encourage Russia to attack a NATO country, Tuberville said he was “not getting in that convers…

Republicans sidestep criticizing Trump’s attack on NATO
The former president’s remarks on not defending NATO allies against Russia were met with a mixture of pushback, defensiveness and silence on Capitol Hill.

Ad's funder puts money toward the Alliance Defending Freedom: "ADF has joined with like-minded organizations in Europe in support of the forced sterilization of transgender individuals and has represented numerous anti-#LGBTQ plaintiffs in pivotal legal battles for LGBTQ rights. The organization has also been a large force behind the anti-abortion movement."
Billion-dollar ad …

Billion-dollar ad campaign fueled by anti-LGBTQ+ forces returns to Super Bowl to promote Jesus - LGBTQ Nation
The campaign preaches love, yet it's major funders also back a prominent anti-LGBTQ+ hate group.

This https://arxiv.org/abs/2402.12168 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning
Recently, various parameter-efficient fine-tuning (PEFT) strategies for application to language models have been proposed and successfully implemented. However, this raises the question of whether PEFT, which only updates a limited set of model parameters, constitutes security vulnerabilities when confronted with weight-poisoning backdoor attacks. In this study, we show that PEFT is more susceptible to weight-poisoning backdoor attacks compared to the full-parameter fine-tuning method, with pre…

This https://arxiv.org/abs/2402.08983 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding
As large language models (LLMs) become increasingly integrated into real-world applications such as code generation and chatbot assistance, extensive efforts have been made to align LLM behavior with human values, including safety. Jailbreak attacks, aiming to provoke unintended and unsafe behaviors from LLMs, remain a significant/leading LLM safety threat. In this paper, we aim to defend LLMs against jailbreak attacks by introducing SafeDecoding, a safety-aware decoding strategy for LLMs to ge…

This https://arxiv.org/abs/2312.14197 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models
The integration of large language models (LLMs) with external content has enabled more up-to-date and wide-ranging applications of LLMs, such as Microsoft Copilot. However, this integration has also exposed LLMs to the risk of indirect prompt injection attacks, where an attacker can embed malicious instructions within external content, compromising LLM output and causing responses to deviate from user expectations. To investigate this important but underexplored issue, we introduce the first be…

Raiders' Maxx Crosby reveals interesting reason for backing Chiefs in Super Bowl LVIII https://www.yardbarker.com/nfl/articles/raiders_maxx_crosby_reveals_interesting_reason_for_backing_c…

Raiders' Maxx Crosby reveals interesting reason for backing Chiefs in Super Bowl LVIII
The Las Vegas Raiders and Kansas City Chiefs rivalry goes back to the days of the American Football League, so it stands to reason that most players representing the silver and black would be rooting against the defending champions in Super Bowl LVIII.  Raiders star pass-rusher Maxx Crosby isn't most people, and on Wednesday, he revealed on NFL Network's "Super Bowl Live" that he wants the Chiefs to hoist another Lombardi Trophy. However, his reasoning is bound to fire Raider Nation up.  "Sel…

SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding
Zhangchen Xu, Fengqing Jiang, Luyao Niu, Jinyuan Jia, Bill Yuchen Lin, Radha Poovendran
https://arxiv.org/abs/2402.08983

SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding
As large language models (LLMs) become increasingly integrated into real-world applications such as code generation and chatbot assistance, extensive efforts have been made to align LLM behavior with human values, including safety. Jailbreak attacks, aiming to provoke unintended and unsafe behaviors from LLMs, remain a significant/leading LLM safety threat. In this paper, we aim to defend LLMs against jailbreak attacks by introducing SafeDecoding, a safety-aware decoding strategy for LLMs to ge…

'Disinformation': Greene pushes pro-Putin propaganda in 'war on Christianity' rant - Alternet.org
https://www.alternet.org/disinformation-greene-putin/

'Disinformation': Greene pushes pro-Putin propaganda in 'war on Christianity' rant
After two top Republican House committee chairmen said Russian propaganda has made its way into Congress and “infected” some in the GOP base, U.S. Rep. Marjorie Taylor Greene (R-GA) spread demonstrably false pro-Russia talking points about a “war on Christianity” while defending and promoting Presid...

This https://arxiv.org/abs/2309.02705 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Certifying LLM Safety against Adversarial Prompting
Large language models (LLMs) are vulnerable to adversarial attacks that add malicious tokens to an input prompt to bypass the safety guardrails of an LLM and cause it to produce harmful content. In this work, we introduce erase-and-check, the first framework for defending against adversarial prompts with certifiable safety guarantees. Given a prompt, our procedure erases tokens individually and inspects the resulting subsequences using a safety filter. Our safety certificate guarantees that har…

This https://arxiv.org/abs/2403.05030 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Defending Against Unforeseen Failure Modes with Latent Adversarial Training
AI systems sometimes exhibit harmful unintended behaviors post-deployment. This is often despite extensive diagnostics and debugging by developers. Minimizing risks from models is challenging because the attack surface is so large. It is not tractable to exhaustively search for inputs that may cause a model to fail. Red-teaming and adversarial training (AT) are commonly used to make AI systems more robust. However, they have not been sufficient to avoid many real-world failure modes that differ…

#Trump pissing away more money: After losing defamation lawsuit, British court orders him to pay Steele firm's legal fees - National Zero

StruQ: Defending Against Prompt Injection with Structured Queries
Sizhe Chen, Julien Piet, Chawin Sitawarin, David Wagner
https://arxiv.org/abs/2402.06363 …

StruQ: Defending Against Prompt Injection with Structured Queries
Recent advances in Large Language Models (LLMs) enable exciting LLM-integrated applications, which perform text-based tasks by utilizing their advanced language understanding capabilities. However, as LLMs have improved, so have the attacks against them. Prompt injection attacks are an important threat: they trick the model to deviate from the original application's instructions and instead follow user directives. These attacks rely on the LLM's ability to follow instructions and inability to s…

This https://arxiv.org/abs/2312.14197 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models
The integration of large language models (LLMs) with external content has enabled more up-to-date and wide-ranging applications of LLMs, such as Microsoft Copilot. However, this integration has also exposed LLMs to the risk of indirect prompt injection attacks, where an attacker can embed malicious instructions within external content, compromising LLM output and causing responses to deviate from user expectations. To investigate this important but underexplored issue, we introduce the first be…

Defending Against Unforeseen Failure Modes with Latent Adversarial Training
Stephen Casper, Lennart Schulze, Oam Patel, Dylan Hadfield-Menell
https://arxiv.org/abs/2403.05030

Defending Against Unforeseen Failure Modes with Latent Adversarial Training
AI systems sometimes exhibit harmful unintended behaviors post-deployment. This is often despite extensive diagnostics and debugging by developers. Minimizing risks from models is challenging because the attack surface is so large. It is not tractable to exhaustively search for inputs that may cause a model to fail. Red-teaming and adversarial training (AT) are commonly used to make AI systems more robust. However, they have not been sufficient to avoid many real-world failure modes that differ…

Defending Against Unforeseen Failure Modes with Latent Adversarial Training
Stephen Casper, Lennart Schulze, Oam Patel, Dylan Hadfield-Menell
https://arxiv.org/abs/2403.05030

Defending Against Unforeseen Failure Modes with Latent Adversarial Training
AI systems sometimes exhibit harmful unintended behaviors post-deployment. This is often despite extensive diagnostics and debugging by developers. Minimizing risks from models is challenging because the attack surface is so large. It is not tractable to exhaustively search for inputs that may cause a model to fail. Red-teaming and adversarial training (AT) are commonly used to make AI systems more robust. However, they have not been sufficient to avoid many real-world failure modes that differ…

This https://arxiv.org/abs/2201.13354 has been replaced.
link: https://scholar.google.com/scholar?q=a

Graph Set-colorings And Hypergraphs In Topological Coding
In order to make more complex number-based strings from topological coding for defending against the intelligent attacks equipped with quantum computing and providing effective protection technology for the age of quantum computing, we will introduce set-colored graphs admitting set-colorings that has been considerable cryptanalytic significance, and especially related with hypergraphs. We use the set-coloring of graphs to reflect the intersection of elements, and add other constraint requireme…

This https://arxiv.org/abs/2201.13354 has been replaced.
link: https://scholar.google.com/scholar?q=a

Graph Set-colorings And Hypergraphs In Topological Coding
In order to make more complex number-based strings from topological coding for defending against the intelligent attacks equipped with quantum computing and providing effective protection technology for the age of quantum computing, we will introduce set-colored graphs admitting set-colorings that has been considerable cryptanalytic significance, and especially related with hypergraphs. We use the set-coloring of graphs to reflect the intersection of elements, and add other constraint requireme…

This https://arxiv.org/abs/2402.17092 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

An Innovative Information Theory-based Approach to Tackle and Enhance The Transparency in Phishing Detection
Phishing attacks have become a serious and challenging issue for detection, explanation, and defense. Despite more than a decade of research on phishing, encompassing both technical and non-technical remedies, phishing continues to be a serious problem. Nowadays, AI-based phishing detection stands out as one of the most effective solutions for defending against phishing attacks by providing vulnerability (i.e., phishing or benign) predictions for the data. However, it lacks explainability in te…

This https://arxiv.org/abs/2402.17092 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

An Innovative Information Theory-based Approach to Tackle and Enhance The Transparency in Phishing Detection
Phishing attacks have become a serious and challenging issue for detection, explanation, and defense. Despite more than a decade of research on phishing, encompassing both technical and non-technical remedies, phishing continues to be a serious problem. Nowadays, AI-based phishing detection stands out as one of the most effective solutions for defending against phishing attacks by providing vulnerability (i.e., phishing or benign) predictions for the data. However, it lacks explainability in te…