30
Smart search ranks by meaning as well as keywords (one row per story, last 45 days).
Rare Not Random Using Token Efficiency for Secrets Scanning
Researcher proposes token efficiency (string length divided by BPE token count) as a better post-regex filter than entropy for secrets scanning, validated on CredData.
The post explores whether Byte-Pair Encoding tokenization can replace Shannon entropy as the primary filter for candidate secrets captured by regex in tools like Gitleaks. It defines 'token efficiency' as string length divided by token count under the cl100k_base tokenizer; secret-like strings such as GitHub tokens tokenize into many small tokens and score low, while natural text scores high. Evaluating labeled secrets from the CredData dataset shows a usable separation, with roughly 2.5 suggested as a minimum cutoff versus Gitleaks' 3.5 entropy threshold. The technique is positioned as a post-regex filtering step rather than a standalone detector.
45
F5 enhances AI Gateway to control AI costs, access, and security
F5 integrated AI Gateway into its AI Security Platform, adding model routing, MCP governance, and guardrails, claiming up to 60% token spend reduction.
F5 announced AI Gateway enhancements combining a Model Gateway for cost optimization, an MCP Gateway for agent-to-tool access control, and AI Guardrails for prompt and response inspection. The company cited its 2026 State of Application Strategy Report finding 77% of organizations now treat inference as their dominant AI activity and manage an average of seven AI models. F5 claims smart routing, semantic caching, and GPU-aware load balancing can cut token spend by up to 60% without application changes. The gateway enforces budgets, model routing policies, and agent access controls centrally across SaaS, hybrid SaaS, and hybrid multicloud deployments, with air-gapped support planned.
30
LexFlip: A Dissociation Diagnostic for Legal Meaning Preservation Metrics
LexFlip releases 373 minimal perturbations of Quebec statutory French that reverse legal force while preserving tokens, exposing weaknesses in embedding-based meaning preservation metrics.
LexFlip provides 373 minimal perturbations of Quebec statutory French that reverse legal force while preserving 0.93 of tokens, creating dissociation items that break monotone token-overlap metric validation. The seven embedding and BERTScore metrics tested register only 0.022-0.039 of their identical-to-unrelated range on these edits, versus 0.670 for bidirectional NLI. Against FrJudge, with a measured human ceiling of r=0.597, a bare length feature outscores every semantic metric tested.
22
Salesforce and Nvidia’s new reasoning model is everything the AI labs should fear
Salesforce and Nvidia launch Koa, Salesforce's first reasoning model, built on Nvidia's open-weight Nemotron and post-trained on synthetic sales and support data.
Salesforce announced Koa at Dreamforce, its first reasoning model, built on Nvidia's open-weight Nemotron and post-trained with synthetic data mimicking sales and customer-support scenarios rather than real customer data. Koa will be offered through the Agentforce platform's AI gateway as a cheaper, token-efficient alternative to closed frontier models like Claude and ChatGPT for enterprise tasks. Salesforce simultaneously announced a ClaudeForce partnership with Anthropic keeping customer data inside Salesforce's infrastructure.
68
Studying Image Tokenizers as Visual Languages in Unified Multimodal Models
A controlled autoregressive testbed shows validation losses must be analyzed per task, and image tokenizer choice affects joint multimodal text modeling.
Researchers built a pure-autoregressive testbed to study image tokenizers as the 'visual language' of unified multimodal models, tracking task-specific validation losses during multimodal continual pretraining across text, image, text-to-image (T2I), and image-to-text (I2T) prediction. They found that losses exhibit distinct scaling behavior per task and rank tokenizers differently, and that I2T loss over a shared text vocabulary gives a more consistent loss–performance signal than T2I loss. Better reconstruction does not necessarily yield lower task-specific losses or stronger downstream performance, and tokenizer choice can affect text modeling under joint optimization. Case studies examine the discriminator, semantic supervision, and vocabulary size design axes.
40
Airrived adds Agentic Observability to track AI agent actions and risks
Airrived launches Agentic Observability to give enterprises end-to-end visibility into AI agent actions, permissions, data flows, and costs.
Airrived announced Agentic Observability, an expansion of its enterprise Agentic OS that traces the full agentic lifecycle from enterprise data ingestion through agent reasoning to business outcomes. The platform surfaces each agent's creator, owner, permissions, permitted actions, and human-in-the-loop approval requirements, and tracks movement of PII, PCI, and PHI across agentic workflows. It also adds token- and model-consumption tracking to turn AI spending into measurable AI FinOps.
18
Studying Image Tokenizers as Visual Languages in Unified Multimodal Models
A controlled pure-autoregressive testbed shows task-specific validation losses rank image tokenizers differently, with I2T loss the most consistent signal.
Researchers built a controlled pure-autoregressive testbed and tracked task-specific validation losses during multimodal continual pretraining across text, image, text-to-image (T2I), and image-to-text (I2T) prediction. They find losses should be analyzed per task because they exhibit distinct scaling behavior and rank tokenizers differently, and that the loss-performance relationship depends on the predicted token space. I2T loss, computed over a shared text vocabulary, correlates consistently with both generation and visual understanding performance after supervised finetuning. Case studies revisit the discriminator, semantic supervision, and vocabulary size as tokenizer design axes.
30
DataGrout helps enterprises control AI usage, governance and LLM costs
SelectHub launched DataGrout, an LLM cost and governance platform combining dynamic context pruning, symbolic inference and MCP gating, claiming about 60% token reduction.
SelectHub launched DataGrout, an LLM inference optimization and AI governance platform combining dynamic context pruning, a symbolic inference layer, and an MCP gateway with per-call token and cost monitoring. Early tests claim roughly 60% token reduction on data-intensive ERP and CRM integration tasks without accuracy loss. The platform connects via its Conduit SDK, MCP or JSON-RPC, offers its own MCP servers for SaaS apps such as Salesforce, SAP and ServiceNow, and supports bring-your-own-key LLM access or gateways like Amazon Bedrock and Kong.
25
Molecular Déjà Vu: Digit-Level Retrieval of Published Values in Frontier Language Models
Audit of 22 frontier models finds widespread verbatim retrieval of published molecular property values, with higher reasoning increasing recall of memorized numbers.
An arXiv audit tests 22 frontier LLMs across 12 molecular regression benchmarks for verbatim retrieval of published values. More than 50% of the LLMs show verbatim retrieval on five datasets, and identical experiments are flagged 89% more often at a high reasoning level than at the lowest one. Suppressing retrieval moves model prediction errors closer together in relative terms, suggesting predictive capability is not determined solely by memorized values.
42
Right Tool, Right Job: Native-Language Evaluation, Tokenizer Sensitivity, and Methodological Findings from a French-Only BabyLM
French BabyLM entry METRON-FR (125M GPT-2, 92.47M words) shows tokenizer artifacts dominate child-scale zero-shot evaluation; proposes standard diagnostics.
METRON-FR is a 125M-parameter GPT-2 pretrained on 92.47M French words, submitted to the BabyLM 2026 Strict track, scoring 85.97% on the native Quebec-French QFrBLiMP benchmark and 62.80% on the BabyLM-weighted leaderboard. A cross-lingual GLUE protocol combining French task-data translation with rank-16 LoRA shows relational tasks gain while world-knowledge tasks regress. Bilingual Lexicon Induction reaches p@1 of 68.84%, 18x above chance, and ablations show single-token zero-shot scoring is dominated by tokenizer and template artifacts at child scale.
12
IndicTriMix: Developing Language Identification Datasets and Models for Tri-Language Code-Mixing
Researchers release IndicTriMix benchmarks and fine-tuned MuRIL and XLM-RoBERTa models for token-level language identification in tri-language code-mixed text.
The paper formulates token-level language identification in code-mixed text as a sequence labeling task and fine-tunes MuRIL and XLM-RoBERTa transformer models for Indian languages. It evaluates on Hindi, Gujarati, and Bengali configurations with manually annotated test sets and proposes two code-mixed generation approaches using parallel trilingual sentences. A public benchmark, annotated test sets, and fine-tuned models are released for reproducibility.
15
Technical Manual for a Toolkit for Measuring Contextual Individuation in Transformer Language Models
An open methodology toolkit measures whether transformer language models contextualize fixed word forms across domains using bridge forms and layer-wise silhouette analysis.
The manual documents an open toolkit built around 'bridge forms' - identical written words recurring across two or more subject domains with a different sense in each - to test whether transformer language models individuate word occurrences by context beyond the embedding layer. It covers declarative specification of bridge forms, Wikipedia corpus acquisition, occurrence localization, layer-wise representation extraction, domain-pairwise silhouette measurement, and visualization, justifying each choice against failure modes such as sense contamination and subword-tokenization misalignment. It is a methodological and implementation reference and reports no empirical results.
14
LLMs and Contextual Integrity
Bruce Schneier highlights two papers: the CIMemories benchmark shows frontier LLMs leak memory attributes up to 69%, and an RL method reduces inappropriate disclosures.
Bruce Schneier discusses contextual integrity in LLMs, referencing the CIMemories benchmark, which uses synthetic profiles with 100+ attributes per user to test whether models with persistent memory disclose sensitive information appropriately. Evaluation showed frontier models exhibit up to 69% attribute-level violations, with GPT-5's violation rate rising from 0.1% to 9.6% across 40 tasks and reaching 25.1% with repeated prompting, showing unstable leakage behavior. A second paper introduces a reinforcement learning framework trained on a synthetic 700-example dataset that substantially reduces inappropriate disclosure while maintaining task performance, with improvements transferring to the human-annotated PrivacyLens benchmark.
48
Unifying Conformal Language Tasks with In-Context Ensembles
Researchers propose Conformal Relevance, which builds conformal score functions via in-context example curation and ensembling to improve conciseness across seven NLP tasks.
The paper targets NLP tasks like summarization and extractive QA that reduce to retrieving content under coverage and conciseness constraints. Conformal Relevance replaces hand-engineered LLM scoring prompts with curated in-context examples and ensembles, maintaining coverage guarantees while improving conciseness with minimal manual input. The authors demonstrate the framework on seven NLP tasks and contribute theory, including a complementarity condition for when ensembling improves worst-case sentence scores and a saturation bound on ensemble gains.
42
NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction
An 8.9B-parameter latent-space language model using next-concept prediction matches OLMo-3-7B pretraining loss with only 51.3% of the training tokens.
NCP-ArchPreview augments next-token prediction with Next Concept Prediction over a product-quantized concept vocabulary built from hidden states, trained jointly end-to-end. The 8.9B model was trained on 5.73T tokens from the Dolma-3 dataset, the largest latent-space language model demonstration to date. It consumes 51.3% of the tokens to reach OLMo-3-7B's final pretraining loss and outperforms it by 2.45 points on the downstream macro-average, including a 5.99-point GSM8K gain. The learned latent space also enables lightweight domain adaptation via a 17M-parameter VQ module and improves speculative drafting accepted length by 4.17%.
52
Augustinian BabyLM: What Ostensive Definition Can and Cannot Teach a Small Language Model
Study shows visually grounded token embeddings in a small masked LM persist through training and improve object-property knowledge, but escape standard BabyLM benchmarks.
The paper implements ostensive definition for a small DeBERTa masked language model trained on 10M words, seeding visually grounded tokens with embeddings derived from labeled image regions before training. Visual initialization leaves a persistent, seed-replicated advantage on object-property knowledge (COMPS) and a corpus-tailored Visual-Property Swap benchmark covering color, material, size, and shape, but has no effect on most BabyLM grammar benchmarks. Synthetic grounding of previously unseeded words causally transfers the advantage to exactly those words.
20
27.5KB language-agnostic WebGPU syntax highlighter
A developer released gpu-lexer, a 27.5KB language-agnostic syntax highlighter that uses a tiny WebGPU model to label code tokens in the browser.
gpu-lexer splits source into words, whitespace, and symbols, then a small WebGPU model uses local and whole-file context to assign nine token classes, working on languages never seen in training. On held-out files, 12.57% of token labels differ from Shiki, though this measures agreement with Shiki rather than objective correctness. In benchmarks against Shiki 4.4.3, Prism.js, Highlight.js, Sugar High, and Starry Night, it highlighted 10 concatenated copies of three.min.js (5.56M characters) about 10x faster on an Apple M4 Pro in Chrome 152. The author frames it as an experiment, not a grammar-equivalent highlighter.
15
CausalArena: Benchmarking Causal Discovery in the Foundation Model Era
Researchers introduce CausalArena, a unified benchmark revealing that causal discovery rankings shift substantially across structural causal model families and protocols.
The paper presents CausalArena, a unified and evolvable benchmark for causal discovery combining synthetic structural causal models, semantically grounded operational SCMs, formula-grounded scientific SCMs, and public real-world datasets. Experiments across classical, neural, and pretrained causal discovery foundation models show large ranking shifts between benchmark regimes. The authors identify pretraining-evaluation overlap and benchmark diversity as central evaluation challenges.
30
TokenRhythm/NeoHorse-1-4B — new model trending #30 on Hugging Face
TokenRhythm releases NeoHorse-1-4B, an Apache-2.0 agentic fine-tune of Qwen3.5-4B claiming +5.93 benchmark macro-average gain.
NeoHorse-1-4B is a roughly 4B-parameter text-only causal language model post-trained by TokenRhythm from Qwen/Qwen3.5-4B for agent harnesses, tool use, coding, and instruction following. It applies routing-guided curriculum SFT and routing-guided on-policy distillation over execution trajectories as an early prototype toward recursive self-improvement (RSI). The release reports a 64.87 macro average across ten benchmarks versus 58.94 for Qwen3.5-4B (+5.93) and is distributed under Apache-2.0, trending #30 on Hugging Face.
28
When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis
Elo-per-token analysis shows LLM agents' marginal gains drop below independent sampling at scale; parallel sessions beat one long session.
The paper proposes Elo-per-token analysis, using a Bradley-Terry model to measure how agent performance scales with token budget on open-ended tasks with continuous scoring. Across four agents and four benchmarks with sessions up to 100M tokens, agents initially convert tokens to Elo faster than independent sampling but eventually slow below the linear-in-log-compute reference. The authors define a scaling inflection point and show that splitting 100M tokens across parallel sessions on FrontierCS Polyomino Packing gains +264 Elo over one long session and +355 over ten short sessions. Human contestants on shared AtCoder Heuristic Contest tasks improve superlinearly, indicating headroom over current agents.
48
Robust Coverless Linguistic Steganography via Sentence Embedding Space with Global Resynchronization
Researchers propose a coverless steganographic framework encoding messages as hierarchical clustering paths in sentence embedding space with a Global Resynchronization Mechanism for robustness.
An arXiv paper proposes encoding secret messages as hierarchical clustering paths in the sentence embedding space rather than token space, improving decoding stability against word- and sentence-level textual perturbations. A Global Resynchronization Mechanism (GRM) reframes variable-length bitstreams as discrete symbols anchored to semantic subspaces to prevent bit-slippage. Experiments show substantial robustness improvements while maintaining embedding capacity and resistance to statistical analysis.
18
I Am No One: Style-Aware Paraphrasing for Text Anonymization
Prompt-driven style-aware paraphrasing with LLMs cuts authorship attribution F1 by 60-70% while preserving content quality.
The paper proposes a style-aware, prompt-driven anonymization approach using pretrained LLMs to build compact stylistic profiles from minimal samples and rewrite text to suppress identifiable style markers while preserving meaning. It addresses stylometric re-identification risks in anonymized text, including ASR transcripts of meetings and call-center calls where leakage persists after acoustic anonymization. Across blog and review datasets, the approach reduces authorship attribution F1 by 60-70%, substantially outperforming both DP-based and non-DP baselines while maintaining readability.
40
MAxBench: A Multinomial Concept Recovery Benchmark
MAxBench evaluates multinomial concept recovery methods, finding affine subspaces steer most reliably but none consistently beats prompting.
MAxBench is a geometry-agnostic evaluation framework for multinomial concept representations in language models, based on sampling from recovered concept representations. It compares 10 localization methods covering 5 geometry types across 6 concepts and 4 models. Findings show affine subspaces steer more reliably than rank-one or linear subspaces due to better non-zero offsets, manifold steering is competitive where applicable, and no method consistently outperforms prompting.
18
Do speech foundation models really learn words?
Researchers show via residualization that later layers of HuBERT and wav2vec 2.0 encode word identity and semantics independently of phonetic content.
The study argues that discriminative ability on words does not imply specialized word representations, since good word discrimination can be explained by phoneme encoding alone. By partialling out phoneme information using residualization, the authors show that later layers of HuBERT and wav2vec 2.0 encode words with reasonable fidelity independently of local phonetic content. Applying this disentanglement approach enhances higher-order linguistic information in word discovery tasks, informing analysis of speech foundation models used for recognition and speech tokens.
28
Distributed and Private Textual Data Synthesis from Embeddings
Researchers propose a distributed differentially private text synthesis method combining DP summaries and secure protocols, removing the need for a trusted curator.
The paper presents a differential privacy and cryptography co-design for synthesizing textual training data without a trusted curator or tightly synchronized user participation. It releases a one-time DP summary in embedding space, identifying frequent semantic regions and their DP centroids to enable training-free offline text synthesis, with semantic support protection to avoid exposing rare user texts. A custom secure protocol enforces end-to-end DP guarantees over distributed user data. Across four benchmarks the approach achieves utility comparable to the state-of-the-art centralized DP synthesis method.
25
Everything in Moderation: Per-Domain Coverage Optima and Alignment-Resistant Domain Gaps in Multi-Domain Mid-Training
Controlled mid-training experiments on Qwen3-8B-Base find each domain has a 10-40% coverage optimum and domain gaps survive alignment SFT.
Using Qwen3-8B-Base (with a 4B replication) across five semantically rule-disjoint KOR-Bench domains, the authors train 30 data allocations spanning the five-domain simplex at five seeds each. All five domains show interior optima in the moderate 10-40% coverage band, and domain gaps persist after a fixed-budget compensatory SFT pass, which raises 116/120 cells yet bridges 0/240 pairs at a 5% threshold. Zero coverage collapses mid-training-only accuracy, though a FineWeb-Edu-only control shows the collapse is partly generic drift. The results argue mid-training data composition requires principled design rather than reliance on later alignment.
30
Context Engineering Inside the Harness: 4 Mechanisms That Beat Context Overflow and Goal Loss on Long-Horizon Tasks
Survey of four harness mechanisms—context budgeting, compaction, todo-state, and memory—that keep long-horizon LLM agents on task across 200+ tool calls.
The article details how agent harnesses, not larger context windows, solve context overflow and goal loss on long-horizon tasks, citing Chroma's Context Rot report showing 18 LLMs (GPT-4.1, Claude 4, Gemini 2.5, Qwen3) degrade on long inputs. Concrete implementations include LangChain Deep Agents offloading tool responses over 20,000 tokens to the filesystem and truncating old tool calls at 85% window usage, and Claude Code capping auto memory at 25KB while re-reading the 5 most recently modified files after compaction. OpenAI's Responses API now offers server-side compaction via context_management with a standalone /responses/compact endpoint, which Codex uses for long-running coding tasks. Manus reports a roughly 100:1 input-to-output token ratio per ~50-tool-call task, motivating todo.md state recitation to prevent goal drift.
52
Characterizing Language Generation in the Limit: Finite Witnesses and a Separation-Width Hierarch
New work characterizes language generation in the limit via finite witnesses, proves a full separation-width hierarchy, and formalizes all results in Lean.
The paper fully characterizes when language generation in the limit is possible for arbitrary families over a countable universe: each target must admit a finite positive witness such that targets activated by any finite sample share an infinite common intersection. It defines positive separation width and proves every level of the resulting hierarchy occurs, with countable families admitting singleton witnesses and unions of families with infinite common cores requiring unbounded finite witnesses. The characterization, a universal normalization, and a diagonal capture lemma are machine-checked in the Lean proof assistant, with the development maintained on GitHub.
18
MetroLLM-Bench: Evaluating Language Models as Transit Kiosk Runtimes
MetroLLM-Bench is a 955-case benchmark testing language models as transit kiosk tool-calling runtimes across six real metro systems.
The benchmark covers 37-414-station metro systems and eleven task categories including routing, fare calculation, disruptions, accessibility, and adversarial input, with 14 deterministic and 8 semantic scoring components. Of 26 models from six vendors, a PEFT-tuned 4B Qwen 3.5 student scored 91.3 on Tier 1, exceeding GPT-5.6 (90.6/90.0), while Muse Glimmer 30B led the composite ranking. A deterministic rule-based baseline reached 84.6, and PEFT gains over base models shrank from +7.03 points at 2B to -0.91 at 27B.
30
Attention Quantization for Tabular Foundation Models
FP8 quantization of attention queries, keys, and values speeds tabular foundation model inference up to 1.7x with no accuracy loss.
The paper develops an FP8 quantization strategy targeting attention calculations (queries, keys, values) in tabular foundation models, arguing attention matters more than weight or KV cache quantization given their differing size and serving patterns versus LLMs. Aligning quantization error between test rows and training rows proves crucial, since misalignment causes drastic accuracy drops. A Triton kernel using explicit FP8 matrix multiplication achieves up to 1.7x speedup over regular 16-bit kernels, with no relevant accuracy loss on TabPFN-v3 and TabICLv2 across TabArena and BeyondArena benchmarks.
28
Show HN: LLM Attention Visualization
A developer released a browser-based tool that visualizes which past tokens influence each LLM output token using aggregated, value-weighted attention scores.
A Show HN project presents a React application built on Transformers.js that renders per-token attention influence by aggregating attention weights scaled by value-vector magnitudes across all attention heads and layers. To expose internal tensors, the author instrumented the ONNX computation graph, hosted a modified model on Hugging Face, and pre-generated prompts to avoid long model downloads in the browser. Demos with a 600-million-parameter model show how verbatim copying draws heavily on source tokens and how single outputs blend information from multiple phrases.
28
What Else Needs Fixing? Exploring Cost-Effective Test-Time Compute for Revision Propagation in Artifacts Generated Through Conversation
A new benchmark shows LLMs reach 68.3-93% accuracy propagating local revisions across conversationally generated artifacts, with parallel-sample selection most cost-effective.
The paper introduces a benchmark for revision propagation: when users request a local change, LLMs must identify dependencies and update all affected parts of an artifact generated through conversation, where context lives in the chat history. Nine revision methods, including sequential reflection and parallel sampling variants, were evaluated on gpt-oss-20b/120b, gpt-5.4-mini, and qwen3.5-9b/27b/122b. Baselines scored 68.3-93% accuracy, and selecting among three parallel samples via LLM-based or medoid selection improved accuracy by 2.2-9.7% as the most cost-effective test-time compute strategy. Code and dataset are released.
30
PARSER: Read in Parallel, Reason in Depth for Long-Context LLM Agents
PARSER uses parallel reader subagents and an RL-trained lead agent for long-context QA, beating baselines and cutting latency up to 11x.
The PARSER paper decouples reading from reasoning: frozen subagents each read one document chunk in parallel while an RL-optimized lead agent iteratively broadcasts queries and aggregates evidence in scatter-gather rounds. On multi-hop QA with 7K to 896K token contexts, a 4B-backbone PARSER beats the strongest sequential memory baseline by 5.7 points on average and 12.0 points at 896K tokens, and a 9B version surpasses DeepSeek-V4-Pro by 6.3 points. Controlled experiments show robustness to evidence position, order, and distance perturbations, with inference latency reduced by up to 11x.
52
Climate-ModernBERT: Revisiting Corpus Composition for Domain-Adaptive Continued Pretraining
Climate-ModernBERT domain-adapted encoders reach 76.3 average F1 across nine climate benchmarks, 2.8 points above vanilla ModernBERT-Base.
The authors continue pretraining ModernBERT-Base on three climate corpora - academic text, climate-filtered web data, and synthetic documents - and compare joint mixtures against parameter-space merging of specialized checkpoints. The best model achieves 76.3 average F1 across nine climate NLP benchmarks, a 2.8-point improvement over the vanilla baseline. Academic climate corpora provide the strongest adaptation signal, and parameter-space merging outperforms joint multi-source training while preserving complementary corpus information; all variants are released.
22