AIBullisharXiv – CS AI · Mar 127/10
🧠Researchers developed Adaptive Activation Cancellation (AAC), a real-time framework that reduces hallucinations in large language models by identifying and suppressing problematic neural activations during inference. The method requires no fine-tuning or external knowledge and preserves model capabilities while improving factual accuracy across multiple model scales including LLaMA 3-8B.
🏢 Perplexity
AINeutralarXiv – CS AI · Mar 117/10
🧠Researchers have developed ALADIN, a framework for analyzing accuracy-latency trade-offs in AI accelerators for embedded systems. The tool enables evaluation of quantized neural networks without requiring deployment on target hardware, potentially reducing development time and costs for AI chip designers.
AINeutralarXiv – CS AI · Mar 117/10
🧠Research analyzes FP4 quantization sensitivity across different layers in large language models using NVFP4 and MXFP4 formats on Qwen2.5 models. The study finds MLP projection layers are most sensitive to quantization, while attention layers show substantial robustness to FP4 precision reduction.
AIBullisharXiv – CS AI · Mar 117/10
🧠Researchers introduce Efficient Draft Adaptation (EDA), a framework that significantly reduces the cost of adapting draft models for speculative decoding when target LLMs are fine-tuned. EDA achieves superior performance through decoupled architecture, data regeneration, and smart sample selection while requiring substantially less training resources than full retraining.
AIBullisharXiv – CS AI · Mar 97/10
🧠Researchers introduce COLD-Steer, a training-free framework that enables efficient control of large language model behavior at inference time using just a few examples. The method approximates gradient descent effects without parameter updates, achieving 95% steering effectiveness while using 50 times fewer samples than existing approaches.
AIBullisharXiv – CS AI · Mar 97/10
🧠Researchers have developed a new technique called activation steering to reduce reasoning biases in large language models, particularly the tendency to confuse content plausibility with logical validity. Their novel K-CAST method achieved up to 15% improvement in formal reasoning accuracy while maintaining robustness across different tasks and languages.
AIBullisharXiv – CS AI · Mar 56/10
🧠Researchers developed EvoPrune, a new method that prunes visual tokens during the encoding stage of Multimodal Large Language Models (MLLMs) rather than after encoding. The technique achieves 2x inference speedup with less than 1% performance loss on video datasets, addressing efficiency bottlenecks in AI models processing high-resolution images and videos.
AIBullisharXiv – CS AI · Mar 56/10
🧠Researchers introduce OSCAR, a new query-dependent online soft compression method for Retrieval-Augmented Generation (RAG) systems that reduces computational overhead while maintaining performance. The method achieves 2-5x speed improvements in inference with minimal accuracy loss across LLMs from 1B to 24B parameters.
🏢 Hugging Face
AIBullisharXiv – CS AI · Mar 57/10
🧠Researchers introduce Multi-Sequence Verifier (MSV), a new technique that improves large language model performance by jointly processing multiple candidate solutions rather than scoring them individually. The system achieves better accuracy while reducing inference latency by approximately half through improved calibration and early-stopping strategies.
AIBullisharXiv – CS AI · Mar 47/104
🧠Researchers propose 'best-of-∞' approach for large language models that uses majority voting with infinite samples, achieving superior performance but requiring infinite computation. They develop an adaptive generation scheme that dynamically selects the optimal number of samples based on answer agreement and extend the framework to weighted ensembles of multiple LLMs.
AIBullisharXiv – CS AI · Mar 47/102
🧠Researchers developed a new channel-adaptive AI algorithm that maximizes inference throughput in 6G edge computing networks by dynamically adjusting computational complexity based on channel conditions. The system uses integrated communication and computation (IC²) to optimize both feature compression and model complexity for mobile edge inference.
AIBullisharXiv – CS AI · Mar 47/103
🧠Nightjar is a new adaptive speculative decoding framework for large language models that dynamically adjusts to system load conditions. It achieves 27.29% higher throughput and up to 20.18% lower latency by intelligently enabling or disabling speculation based on workload demands.
AIBullisharXiv – CS AI · Mar 47/103
🧠Researchers introduce Energy Landscape Steering (ELS), a new framework that reduces false refusals in AI safety-aligned language models without compromising security. The method uses an external Energy-Based Model to dynamically guide model behavior during inference, improving compliance from 57.3% to 82.6% on safety benchmarks.
AIBullisharXiv – CS AI · Mar 37/104
🧠Researchers developed a new robotic policy framework using dense-jump flow matching with non-uniform time scheduling to address performance degradation in multi-step inference. The approach achieves up to 23.7% performance gains over existing baselines by optimizing integration scheduling during training and inference phases.
AIBullisharXiv – CS AI · Mar 37/105
🧠Researchers developed HierarchicalPrune, a compression framework that reduces large-scale text-to-image diffusion models' memory footprint by 77.5-80.4% and latency by 27.9-38.0% while maintaining image quality. The technique enables billion-parameter AI models to run efficiently on resource-constrained devices through hierarchical pruning and knowledge distillation.
AIBullisharXiv – CS AI · Mar 37/104
🧠Researchers have developed BWCache, a training-free method that accelerates Diffusion Transformer (DiT) video generation by up to 6× through block-wise feature caching and reuse. The technique exploits computational redundancy in DiT blocks across timesteps while maintaining visual quality, addressing a key bottleneck in real-world AI video generation applications.
AIBullisharXiv – CS AI · Mar 37/103
🧠Researchers have developed SageBwd, a trainable INT8 attention mechanism that can match full-precision attention performance during pre-training while quantizing six of seven attention matrix multiplications. The study identifies key factors for stable training including QK-norm requirements and the impact of tokens per step on quantization errors.
AIBullisharXiv – CS AI · Feb 277/106
🧠Researchers developed ViT-Linearizer, a distillation framework that transfers Vision Transformer knowledge into linear-time models, addressing quadratic complexity issues for high-resolution inputs. The method achieves 84.3% ImageNet accuracy while providing significant speedups, bridging the gap between efficient RNN-based architectures and transformer performance.
AIBullisharXiv – CS AI · Feb 277/106
🧠Researchers developed a runtime-reconfigurable bitwise systolic array architecture for multi-precision quantized neural networks on FPGA hardware accelerators. The system achieves 1.3-3.6x speedup on mixed-precision models while supporting higher clock frequencies up to 250MHz, addressing the trade-off between hardware efficiency and inference accuracy.
AIBullisharXiv – CS AI · Feb 277/107
🧠Researchers introduce Spatial Credit Redistribution (SCR), a training-free method that reduces hallucination in vision-language models by 4.7-6.0 percentage points. The technique redistributes attention from dominant visual patches to contextual areas, addressing the spatial credit collapse problem that causes AI models to generate false objects.
AINeutralCrypto Briefing · Jun 236/10
🧠Cerebras has confirmed that OpenAI's GPT 5.4 model runs on its AI infrastructure, with plans to support GPT 5.5 rollout. While the partnership highlights Cerebras' competitive positioning in AI chip infrastructure, the company faces significant concentration risk from its heavy dependence on a single major client.
🏢 OpenAI🧠 GPT-5
AIBullisharXiv – CS AI · Jun 236/10
🧠Researchers introduce AdaMem, an adaptive memory system for LLM agents that learns what information to retain based on individual user preferences rather than storing everything. The method achieves up to 9% QA accuracy improvement while reducing memory bloat, addressing practical constraints of inference costs and finite context windows in production systems.
AIBullisharXiv – CS AI · Jun 236/10
🧠Researchers introduce Denoising Iterative Self-Correction (DISC), a test-time procedure that improves large language model reasoning by treating verification outputs as noisy signals to progressively correct errors across multiple passes. The method demonstrates superior performance over existing correction approaches, achieving 81.6% accuracy on BIG-Bench Mistake with 13x better improvement-to-degradation ratios than Chain-of-Verification.
AIBullisharXiv – CS AI · Jun 236/10
🧠Researchers introduce FOCUS, a training-free method that improves Large Vision-Language Models' ability to process multiple images by masking irrelevant images with noise, preventing visual information from different images from becoming entangled in the model's representations.
AIBullisharXiv – CS AI · Jun 236/10
🧠Researchers propose a training-free caching strategy that accelerates molecular geometry generation in flow matching models by predicting intermediate hidden states, achieving 2-7x speedups without quality degradation. The method is compatible with pretrained models and compounds with existing optimizations, addressing a critical inference bottleneck in computational chemistry workflows.
