https://scatterai.com/brief Plain-English explanations of AI research papers, written for builders. Daily. en https://scatterai.com/brief/2026-07-08 https://scatterai.com/brief/2026-07-08 Wed, 08 Jul 2026 12:00:00 GMT A formal proof shows ColBERT-style MaxSim can replicate any similarity dense or sparse retrieval can express, plus functions they cannot. https://scatterai.com/brief/2026-07-08-awn https://scatterai.com/brief/2026-07-08-awn Wed, 08 Jul 2026 12:00:00 GMT Five papers on closing gaps: sparse attention, hard-prompt RL, tri-mode decoding, semantic caching, and agent self-evolution. https://scatterai.com/brief/2026-07-07 https://scatterai.com/brief/2026-07-07 Tue, 07 Jul 2026 12:00:00 GMT SeKV stores KV cache entries at variable resolution across GPU and CPU, reconstructing fine-grained detail on demand and beating the best compression baseline by 5.9%. https://scatterai.com/brief/2026-07-07-awn https://scatterai.com/brief/2026-07-07-awn Tue, 07 Jul 2026 12:00:00 GMT Five papers on training signals, scaling laws, optimizer selection, embodied generalization, and representation surgery https://scatterai.com/brief/2026-07-06 https://scatterai.com/brief/2026-07-06 Mon, 06 Jul 2026 12:00:00 GMT Training-inference probability mismatch silently corrupts LLM RL post-training. MIPU fixes the actual objective: the inference policy, not the training loss. https://scatterai.com/brief/2026-07-06-awn https://scatterai.com/brief/2026-07-06-awn Mon, 06 Jul 2026 12:00:00 GMT Five papers on evaluation gaps, data quality, and deployment friction across agents, RAG, and edge robotics https://scatterai.com/brief/2026-07-05 https://scatterai.com/brief/2026-07-05 Sun, 05 Jul 2026 12:00:00 GMT A new diagnostic shows that changing only the evaluation harness shifts multi-step agent beliefs, making cross-framework leaderboard comparisons unreliable. https://scatterai.com/brief/2026-07-05-awn https://scatterai.com/brief/2026-07-05-awn Sun, 05 Jul 2026 12:00:00 GMT Stale caches, forgotten RL training, transplantable misalignment, noisy benchmarks, and diffusion LLM serving , five cracks in common assumptions https://scatterai.com/brief/2026-07-04 https://scatterai.com/brief/2026-07-04 Sat, 04 Jul 2026 12:00:00 GMT MemSyco-Bench is the first benchmark isolating memory-induced sycophancy in LLM agents, revealing that retrieved memories systematically corrupt factual reasoning. https://scatterai.com/brief/2026-07-04-awn https://scatterai.com/brief/2026-07-04-awn Sat, 04 Jul 2026 12:00:00 GMT Bounded memory, cheap eval proxies, step-aware RL, training-free diffusion speedup, and non-literal retrieval heads https://scatterai.com/brief/2026-07-03 https://scatterai.com/brief/2026-07-03 Fri, 03 Jul 2026 12:00:00 GMT PAW compiles natural-language function specs into tiny local adapters, matching Qwen3-32B performance at 1/50th the memory with no API calls required. https://scatterai.com/brief/2026-07-03-awn https://scatterai.com/brief/2026-07-03-awn Fri, 03 Jul 2026 12:00:00 GMT Five papers on model internals: hybrid layer selection, forgetting myths, weight auditing, VLA pretraining, and trainable memory. https://scatterai.com/brief/2026-07-02 https://scatterai.com/brief/2026-07-02 Thu, 02 Jul 2026 12:00:00 GMT A new proof shows GRPO, Dr. GRPO, and DAPO all reduce to adjusting one scalar: the group standard deviation of per-prompt answer correctness. https://scatterai.com/brief/2026-07-02-awn https://scatterai.com/brief/2026-07-02-awn Thu, 02 Jul 2026 12:00:00 GMT Five papers on making LLM systems faster, smarter, and more honest: routing, training, self-improvement, evaluation, and architecture. https://scatterai.com/brief/2026-07-01 https://scatterai.com/brief/2026-07-01 Wed, 01 Jul 2026 12:00:00 GMT RLMF trains LLMs to accurately express uncertainty by using self-judgment quality as the RL reward signal, beating standard RL by up to 63%. https://scatterai.com/brief/2026-07-01-awn https://scatterai.com/brief/2026-07-01-awn Wed, 01 Jul 2026 12:00:00 GMT Five papers exposing hidden inefficiencies in how LLMs are trained, evaluated, and decoded https://scatterai.com/brief/2026-06-30 https://scatterai.com/brief/2026-06-30 Tue, 30 Jun 2026 12:00:00 GMT Agents-A1 reaches 1T-parameter-level benchmark scores by extending trajectory length and unifying heterogeneous tool domains, not by adding parameters. https://scatterai.com/brief/2026-06-30-awn https://scatterai.com/brief/2026-06-30-awn Tue, 30 Jun 2026 12:00:00 GMT From threshold-free KV compression to Ridge regression beating transformers, five papers questioning the assumptions underneath production ML. https://scatterai.com/brief/2026-06-29 https://scatterai.com/brief/2026-06-29 Mon, 29 Jun 2026 12:00:00 GMT Conversational infill streams partial reasoning from a large model into a small real-time talker mid-utterance, hitting sub-300ms latency without a capability cliff. https://scatterai.com/brief/2026-06-29-awn https://scatterai.com/brief/2026-06-29-awn Mon, 29 Jun 2026 12:00:00 GMT SAE failure modes, two cost-cutting inference papers, static anchors for code agents, and when to skip test execution entirely https://scatterai.com/brief/2026-06-28 https://scatterai.com/brief/2026-06-28 Sun, 28 Jun 2026 12:00:00 GMT Embedding space geometry, not training method, decides whether cosine or rank-based metrics win , and a single variance stat predicts it with 0.95 correlation. https://scatterai.com/brief/2026-06-28-awn https://scatterai.com/brief/2026-06-28-awn Sun, 28 Jun 2026 12:00:00 GMT Memory eviction policy, majority-vote failure modes, semantic join scaling, benchmark fidelity gaps, and latent behavior elicitation https://scatterai.com/brief/2026-06-27 https://scatterai.com/brief/2026-06-27 Sat, 27 Jun 2026 12:00:00 GMT OPID extracts hierarchical skill signals directly from on-policy trajectories, replacing sparse outcome rewards with dense token-level supervision for agentic RL. https://scatterai.com/brief/2026-06-27-awn https://scatterai.com/brief/2026-06-27-awn Sat, 27 Jun 2026 12:00:00 GMT Five papers on where current agent and LLM infrastructure quietly breaks: memory evals, retrieval reasoning, GUI vs. CLI execution, token compression, and free process rewards. https://scatterai.com/brief/2026-06-26 https://scatterai.com/brief/2026-06-26 Fri, 26 Jun 2026 12:00:00 GMT As coding agents grow more capable, generating solutions gets easier. Reliably verifying them is now the harder problem, and no fixed reward function survives it. https://scatterai.com/brief/2026-06-26-awn https://scatterai.com/brief/2026-06-26-awn Fri, 26 Jun 2026 12:00:00 GMT A hard ceiling on multi-model ensembles, a citation failure rate of 15.9%, and three inference/training fixes for practitioners shipping today https://scatterai.com/brief/2026-06-25 https://scatterai.com/brief/2026-06-25 Thu, 25 Jun 2026 12:00:00 GMT New evidence shows reasoning models lock in refusal/compliance before visible thinking begins, with 0.84-0.95 AUROC predictability from the first token. https://scatterai.com/brief/2026-06-25-awn https://scatterai.com/brief/2026-06-25-awn Thu, 25 Jun 2026 12:00:00 GMT Five papers exposing silent failure modes in agents, quantization, and training assumptions practitioners are likely shipping around. https://scatterai.com/brief/2026-06-24 https://scatterai.com/brief/2026-06-24 Wed, 24 Jun 2026 12:00:00 GMT NatureBench tests 10 frontier agents on 90 containerized tasks from Nature-family papers. The best model surpasses published SOTA on just 17.8% of them. https://scatterai.com/brief/2026-06-24-awn https://scatterai.com/brief/2026-06-24-awn Wed, 24 Jun 2026 12:00:00 GMT Five papers tightening the feedback loops that break quietly: agent memory, retrieval training, data mixing, embedding cost, and diffusion eval. https://scatterai.com/brief/2026-06-23 https://scatterai.com/brief/2026-06-23 Tue, 23 Jun 2026 12:00:00 GMT EnterpriseClawBench converts real workplace sessions into 852 reproducible agent tasks, exposing a 0.663 ceiling even for GPT-5.5-powered agents. https://scatterai.com/brief/2026-06-23-awn https://scatterai.com/brief/2026-06-23-awn Tue, 23 Jun 2026 12:00:00 GMT Head-level attention hybridization, premature agent commitment, intermediate-layer decoding, contextual privacy evals, and single-proxy data mixing https://scatterai.com/brief/2026-06-22 https://scatterai.com/brief/2026-06-22 Mon, 22 Jun 2026 12:00:00 GMT GateMem benchmarks shared-memory agents across hospitals, offices, and households, and finds no current method reliably handles governance alongside recall. https://scatterai.com/brief/2026-06-22-awn https://scatterai.com/brief/2026-06-22-awn Mon, 22 Jun 2026 12:00:00 GMT From bias auditing to model merging at scale, five papers that tighten the gap between theory and production. https://scatterai.com/brief/2026-06-21 https://scatterai.com/brief/2026-06-21 Sun, 21 Jun 2026 12:00:00 GMT A 36-month leakage-controlled test shows a 7B RAG forecaster's median IC of +0.154 is largely explained by macro-analog retrieval, not LLM capability. https://scatterai.com/brief/2026-06-21-awn https://scatterai.com/brief/2026-06-21-awn Sun, 21 Jun 2026 12:00:00 GMT Two eval frameworks exposing hidden gaps, two GRPO training fixes, and one materials-science RL environment. https://scatterai.com/brief/2026-06-20 https://scatterai.com/brief/2026-06-20 Sat, 20 Jun 2026 12:00:00 GMT Self-directed robot play before task assignment builds a reusable skill library that lifts downstream performance by up to 20.6 points, no finetuning required. https://scatterai.com/brief/2026-06-20-awn https://scatterai.com/brief/2026-06-20-awn Sat, 20 Jun 2026 12:00:00 GMT Single-frame robot world models, grounded visual reasoning, and retriever-aware RAG query rewriting headline today's five papers https://scatterai.com/brief/2026-06-19 https://scatterai.com/brief/2026-06-19 Fri, 19 Jun 2026 12:00:00 GMT A new auxiliary RL objective forces LLMs to select the context fragment that supports an answer, yielding +2.2% on long-horizon agent benchmarks and +1.8% on visual QA. https://scatterai.com/brief/2026-06-19-awn https://scatterai.com/brief/2026-06-19-awn Fri, 19 Jun 2026 12:00:00 GMT From stateful spatial agents to egocentric pretraining, five papers rethink what the right input signal actually is https://scatterai.com/brief/2026-06-18 https://scatterai.com/brief/2026-06-18 Thu, 18 Jun 2026 12:00:00 GMT Clamping SAE features suppresses one path to harmful behavior, not the behavior itself. Models recover through the unexplained reconstruction residual. https://scatterai.com/brief/2026-06-18-awn https://scatterai.com/brief/2026-06-18-awn Thu, 18 Jun 2026 12:00:00 GMT Five papers tightening the screws on RL training stability, rollout speed, agent evaluation, and multicultural system design https://scatterai.com/brief/2026-06-17 https://scatterai.com/brief/2026-06-17 Wed, 17 Jun 2026 12:00:00 GMT ProCUA-SFT shows AgentNet causes negative transfer in CUA fine-tuning, while 3.1M synthetic steps lift OSWorld from 26.3% to 45.0%. https://scatterai.com/brief/2026-06-17-awn https://scatterai.com/brief/2026-06-17-awn Wed, 17 Jun 2026 12:00:00 GMT Five papers on training signals, inference architecture, and efficiency trade-offs across diffusion LLMs, hybrid attention, and coding agents https://scatterai.com/brief/2026-06-16 https://scatterai.com/brief/2026-06-16 Tue, 16 Jun 2026 12:00:00 GMT WebStep's 1,800-task benchmark reveals that agents scoring identically on task success diverge sharply on where and how they fail mid-workflow. https://scatterai.com/brief/2026-06-16-awn https://scatterai.com/brief/2026-06-16-awn Tue, 16 Jun 2026 12:00:00 GMT From KV cache fragmentation fixes to SAE feature instability, five papers that quietly reshape how practitioners build and trust AI systems https://scatterai.com/brief/2026-06-15 https://scatterai.com/brief/2026-06-15 Mon, 15 Jun 2026 12:00:00 GMT MedMisBench shows LLM accuracy drops from 71.1% to 38.0% when misleading context is injected, exposing a structural gap in medical AI evaluation. https://scatterai.com/brief/2026-06-15-awn https://scatterai.com/brief/2026-06-15-awn Mon, 15 Jun 2026 12:00:00 GMT Five papers on where current assumptions break: RL credit, LoRA tuning, inference depth, AI peer review, and video retrieval. https://scatterai.com/brief/2026-06-14 https://scatterai.com/brief/2026-06-14 Sun, 14 Jun 2026 12:00:00 GMT Activation monitors trained on base models degrade sharply after LoRA fine-tuning but survive quantization, exposing a silent gap in most production safety stacks. https://scatterai.com/brief/2026-06-14-awn https://scatterai.com/brief/2026-06-14-awn Sun, 14 Jun 2026 12:00:00 GMT Memory placement failures, formal proof shortcuts, answer instability, federated LoRA aggregation, and 207k coding agent trajectories