LLMs & Texto — Radar de IA

On the Generalization of Steering Vectors for Chain-of-Thought Faithfulness

arXiv:2607.29062v1 Announce Type: new Abstract: Model capabilities have improved in large part due to scaling chain of thought. This has been a promising development for AI safety--where models verbalize their reasoning, it is possible to monitor it. However, in some cases, models do not verbalize important steps in their reasoning process. For example, models prompted with a cue suggesting the incorrect answer may fail to acknowledge that cue, even when it appears instrumental to their conclusi...

03.08.2026

Blog LLMs & Texto

Faster but Different: Diagnosing and Controlling Content Drift in Accelerated Multimodal Diffusion Language Models

arXiv:2607.29079v1 Announce Type: new Abstract: Training-free acceleration makes diffusion-based multimodal large language models (dMLLMs) more deployable, but it may silently change generated content. We study this serving-time consistency problem on 300 real images, comparing Fast-dLLM outputs with the same model's unaccelerated outputs. Across the mild parallelism induced in our long-form setting (1.05--1.25 committed tokens per step), confidence-threshold tuning changes decoding behavior but...

03.08.2026

Blog Multimodal

ZeroR@CHiPSAL 2026: Two-Stage Vision-Language Adaptation with Contrastive Learning for Nepali Meme Classification

arXiv:2607.28637v1 Announce Type: new Abstract: This paper presents our system for the CHiPSAL 2026 shared task on multimodal hate speech and sentiment detection in Nepali memes. We address both subtasks: binary hate speech classification and three-class sentiment analysis. Our approach adapts the Robust Adaptation of Hateful Meme Detection (RA-HMD) framework using Qwen3-VL-8B-Instruct, a state-of-the-art vision-language model with native Devanagari support. We employ a two-stage training pipeli...

03.08.2026

Blog LLMs & Texto

MMShopBench: A Real-Log Benchmark for Multimodal, Multi-Turn Shopping Agents

arXiv:2607.29002v1 Announce Type: new Abstract: Online shoppers increasingly turn to AI shopping assistants, using images and multi-turn dialogue to express and refine product needs that are difficult to articulate in text alone. However, existing benchmarks largely rely on text-only or synthetic requests, underrepresenting complex real-world shopping requirements jointly expressed through images and language. We introduce MMShopBench, the first real-log benchmark for multimodal, multi-turn shop...

03.08.2026

Blog LLMs & Texto

Are the Financial Reasoning from LLMs Credible? A Real World Test over Long-Horizon Statements

arXiv:2607.28661v1 Announce Type: new Abstract: Do Large Language Models (LLMs) possess genuine structural reasoning, or merely rely on surface-level pattern matching? The financial domain, demanding numerical precision and multi-step logic over long contexts, is an ideal testbed. Existing benchmarks fail to capture real-world industrial complexity, predominantly relying on multiple-choice questions or single-hop QA over cropped tables while ignoring intricate cross-statement dynamics and tempor...

03.08.2026

Blog LLMs & Texto

Tokenizer-Agnostic Engram Module

arXiv:2607.29065v1 Announce Type: new Abstract: Deepseek's Engram, a conditional memory module, was introduced to trade-off storage versus reasoning in large language models. However, the module relies on token-level $N$-gram hashing for Engram embedding lookup, introducing a tight coupling to the tokenizer used: a model with a different tokenizer would have to train its own Engram embeddings from scratch. To improve the reusability of Engram embeddings, we propose a change to the hashing routin...

03.08.2026

Blog LLMs & Texto

PARALLEL: A Prefrontal-Aligned Reinforcement inspired Approach for Language-Model Learning under Explicit Limits

arXiv:2607.28982v1 Announce Type: new Abstract: Recent language models achieve strong performance across a variety of tasks, but conventional adaptation applies updates uniformly across training samples regardless of their local update benefit. We propose PARALLEL, a prefrontal-aligned reinforcement inspired approach for language-model learning. Inspired by the complementary roles of goal-related and uncertainty-related control, PARALLEL represents these forms of information as separate controll...

03.08.2026

Blog LLMs & Texto

Self-Supervised Skill Optimization

arXiv:2607.28777v1 Announce Type: new Abstract: Agent skills provide frozen large language model (LLM) agents with reusable procedural guidance, and recent work shows that such skills can be optimized with ground-truth (GT) feedback. Many applications, however, lack GT labels, task scores, rewards, or reliable task-specific evaluators. We therefore introduce Self-Supervised Skill Optimization (SSO), a comparative framework that learns a reusable skill from unlabeled task instances alone. At each...

03.08.2026

Blog LLMs & Texto

SafeNexus: Discovering and Steering Modality-Universal Safety Neurons in MLLMs

arXiv:2607.28969v1 Announce Type: new Abstract: Although Large Language Models (LLMs) have demonstrated promising safety performance, extending them to Multimodal Large Language Models (MLLMs) exposes a significant gap between expanded multimodal capabilities and existing safety mechanisms. Current defenses remain predominantly confined to specific modal settings, thereby limiting their robustness against broader cross-modal threats. To bridge this gap, we introduce SafeNexus, a cross-modal safe...

03.08.2026

Blog LLMs & Texto

Overcoming the Weakest-Link Effect in LLM-Driven Program Optimization via Heterogeneous Edit Recombination

arXiv:2607.28947v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to solve complex problems by searching over program space, offering a general paradigm for scientific problems that can be naturally represented and solved as programs. Despite recent progress, identifying effective optimization directions for a candidate program remains challenging. By analogy with automatic differentiation, existing methods typically guide the search using a textual ``gradient'':...

03.08.2026

Blog LLMs & Texto

Multi-Agent Planning with Spatio-Temporal and Topological Constraints using STL-GO

arXiv:2607.28679v1 Announce Type: new Abstract: Multi-agent planning problems arise in a variety of engineering applications, such as multi-robot wildfire fighting and unmanned aerial inspection in factories. A particular challenge is the existence of spatio-temporal (i.e., when and/or where an agent should do what) and topological constraints (i.e., how agents should interact), as typically formalized via the notion of graphs. Over the last years, various frameworks have been proposed that can ...

03.08.2026

Blog LLMs & Texto

TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs

arXiv:2607.28640v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) should generate consistent responses given semantically equivalent inputs across modalities. However, we observe a systematic discrepancy in model predictions under such cross-modal variations. Specifically, we define the modality gap as the difference in model performance under semantically equivalent textual and multimodal inputs. We introduce TokenSwap, a method that constructs such inputs by replacing te...

03.08.2026