Tootfinder

Emergent Alignment via Competition
Natalie Collina, Surbhi Goel, Aaron Roth, Emily Ryu, Mirah Shi
https://arxiv.org/abs/2509.15090 https://arxiv.org/pdf/25…

Emergent Alignment via Competition
Aligning AI systems with human values remains a fundamental challenge, but does our inability to create perfectly aligned models preclude obtaining the benefits of alignment? We study a strategic setting where a human user interacts with multiple differently misaligned AI agents, none of which are individually well-aligned. Our key insight is that when the users utility lies approximately within the convex hull of the agents utilities, a condition that becomes easier to satisfy as model diversi…

🇺🇦 #NowPlaying on KEXP's #AfternoonShow
Alexandra Savior:
🎵 The Archer
#AlexandraSavior
https://alexandrasavior.bandcamp.com/album/the-archer
https://open.spotify.com/track/2L5aj2upg0rqORcCnPoCiX

I’m very disappointed in Carney, he’s actually a Conservative. I don’t think Poilievre would have been better, but at least I wouldn’t have been lied to.
#CdnPoli
https://mstdn.ca/@AlisonCreekside/1148

Alison Creekside (@AlisonCreekside@mstdn.ca)
Attached: 2 images What a difference winning an election makes ... April 4 : Carney pledges $150M boost to 'underfunded' CBC July 16: Suggest your own 15% spending cuts, Carney government tells CBC #cdnpoli

MAGA Megachurch Pastor Named In Lawsuits Alleging That He Failed To Prevent Molestations By Fellow Pastor - Joe.My.God.
https://www.joemygod.com/2025/09/maga-megachurch-pastor-named-in-lawsuits-alleging-that-he-failed-to-prevent-molestations-by-fellow-pastor/

Beyond Surface Alignment: Rebuilding LLMs Safety Mechanism via Probabilistically Ablating Refusal Direction
Yuanbo Xie, Yingjie Zhang, Tianyun Liu, Duohe Ma, Tingwen Liu
https://arxiv.org/abs/2509.15202

Beyond Surface Alignment: Rebuilding LLMs Safety Mechanism via Probabilistically Ablating Refusal Direction
Jailbreak attacks pose persistent threats to large language models (LLMs). Current safety alignment methods have attempted to address these issues, but they experience two significant limitations: insufficient safety alignment depth and unrobust internal defense mechanisms. These limitations make them vulnerable to adversarial attacks such as prefilling and refusal direction manipulation. We introduce DeepRefusal, a robust safety alignment framework that overcomes these issues. DeepRefusal forc…

Decoupled Proxy Alignment: Mitigating Language Prior Conflict for Multimodal Alignment in MLLM
Chenkun Tan, Pengyu Wang, Shaojun Zhou, Botian Jiang, Zhaowei Li, Dong Zhang, Xinghao Wang, Yaqian Zhou, Xipeng Qiu
https://arxiv.org/abs/2509.14735

Decoupled Proxy Alignment: Mitigating Language Prior Conflict for Multimodal Alignment in MLLM
Multimodal large language models (MLLMs) have gained significant attention due to their impressive ability to integrate vision and language modalities. Recent advancements in MLLMs have primarily focused on improving performance through high-quality datasets, novel architectures, and optimized training strategies. However, in this paper, we identify a previously overlooked issue, language prior conflict, a mismatch between the inherent language priors of large language models (LLMs) and the lan…

Towards Rationale-Answer Alignment of LVLMs via Self-Rationale Calibration
Yuanchen Wu, Ke Yan, Shouhong Ding, Ziyin Zhou, Xiaoqiang Li
https://arxiv.org/abs/2509.13919 https://…

Towards Rationale-Answer Alignment of LVLMs via Self-Rationale Calibration
Large Vision-Language Models (LVLMs) have manifested strong visual question answering capability. However, they still struggle with aligning the rationale and the generated answer, leading to inconsistent reasoning and incorrect responses. To this end, this paper introduces the Self-Rationale Calibration (SRC) framework to iteratively calibrate the alignment between rationales and answers. SRC begins by employing a lightweight "rationale fine-tuning" approach, which modifies the model's respons…

Language models align with brain regions that represent concepts across modalities
Maria Ryskina, Greta Tuckute, Alexander Fung, Ashley Malkin, Evelina Fedorenko
https://arxiv.org/abs/2508.11536

Language models align with brain regions that represent concepts across modalities
Cognitive science and neuroscience have long faced the challenge of disentangling representations of language from representations of conceptual meaning. As the same problem arises in today's language models (LMs), we investigate the relationship between LM--brain alignment and two neural metrics: (1) the level of brain activation during processing of sentences, targeting linguistic processing, and (2) a novel measure of meaning consistency across input modalities, which quantifies how consiste…

Consistent View Alignment Improves Foundation Models for 3D Medical Image Segmentation
Puru Vaish, Felix Meister, Tobias Heimann, Christoph Brune, Jelmer M. Wolterink
https://arxiv.org/abs/2509.13846

Consistent View Alignment Improves Foundation Models for 3D Medical Image Segmentation
Many recent approaches in representation learning implicitly assume that uncorrelated views of a data point are sufficient to learn meaningful representations for various downstream tasks. In this work, we challenge this assumption and demonstrate that meaningful structure in the latent space does not emerge naturally. Instead, it must be explicitly induced. We propose a method that aligns representations from different views of the data to align complementary information without inducing false…

FedSSG: Expectation-Gated and History-Aware Drift Alignment for Federated Learning
Zhanting Zhou, Jinshan Lai, Fengchun Zhang, Zeqin Wu, Fengli Zhang
https://arxiv.org/abs/2509.13895

FedSSG: Expectation-Gated and History-Aware Drift Alignment for Federated Learning
Non-IID data and partial participation induce client drift and inconsistent local optima in federated learning, causing unstable convergence and accuracy loss. We present FedSSG, a stochastic sampling-guided, history-aware drift alignment method. FedSSG maintains a per-client drift memory that accumulates local model differences as a lightweight sketch of historical gradients; crucially, it gates both the memory update and the local alignment term by a smooth function of the observed/expected p…