iGEN
Visit IGEN World Explore IGEN Expo
EXPLORE UPGRADE PLANS
BREAKING
Werner Enterprises Posts Highest Revenue Per Truck Growth in One-Way Segment in a Decade CMA CGM and Stonepeak Launch United Ports LLC in $2.4 Billion Terminal Joint Venture UPS shift away from Amazon shows bigger payoff Lanesurf: 62% of Loads Get Vetted Carrier Offers Before Brokers Arrive India-China Border Trade Via Lipulekh Resumes Aug 1; China Permits 20 Traders Geopolitics Drives CMA CGM Q2 Profit Surge of 42% as Volumes and Rates Climb Benchmark Diesel Price Rises Third Week as Futures Plunge; Spread Hits Record Indian Government Limits Sugar Dealers to 400 Tonnes Stock Until November to Curb Hoarding Tenants signing longer leases for larger warehouses as 3PLs lock in capacity US stock market flat as S&P 500 and Dow barely move, Nasdaq slides over 1% on chip rout Werner Enterprises Posts Highest Revenue Per Truck Growth in One-Way Segment in a Decade CMA CGM and Stonepeak Launch United Ports LLC in $2.4 Billion Terminal Joint Venture UPS shift away from Amazon shows bigger payoff Lanesurf: 62% of Loads Get Vetted Carrier Offers Before Brokers Arrive India-China Border Trade Via Lipulekh Resumes Aug 1; China Permits 20 Traders Geopolitics Drives CMA CGM Q2 Profit Surge of 42% as Volumes and Rates Climb Benchmark Diesel Price Rises Third Week as Futures Plunge; Spread Hits Record Indian Government Limits Sugar Dealers to 400 Tonnes Stock Until November to Curb Hoarding Tenants signing longer leases for larger warehouses as 3PLs lock in capacity US stock market flat as S&P 500 and Dow barely move, Nasdaq slides over 1% on chip rout
Home ›› Technology ›› Ai ›› Llms ›› SoftSkill: Compressing AI Agent Skills into Compact Latent Controls Boosts Accuracy Over Traditional Prompting

SoftSkill: Compressing AI Agent Skills into Compact Latent Controls Boosts Accuracy Over Traditional Prompting

Researchers propose SoftSkill, a method that compresses natural-language agent skills into compact continuous vectors, improving accuracy on benchmarks like LiveMath by 42.1 points over no-skill prompting. The approach uses a frozen backbone and a trainable soft delta, offering a more efficient alternative to traditional Markdown skill files.

iG
iGEN Editorial
July 8, 2026
SoftSkill: Compressing AI Agent Skills into Compact Latent Controls Boosts Accuracy Over Traditional Prompting

Problem: Deploying AI agents with custom skills often relies on lengthy Markdown files that contain instructions, policies, and procedures. These files must be re-interpreted by a frozen language model at inference time, consuming tokens and computation. A team of researchers from an academic institution asks whether a natural-language skill can instead be compressed into a compact continuous object that initializes a latent behavioral prior, dramatically reducing overhead while maintaining or improving performance.

Technical Approach

The proposed method, SoftSkill, keeps the backbone model frozen and tunes a continuous "soft skill" using next-token prediction. According to the paper, the soft skill is a "length-32 prefix" that serves as a latent behavioral prior at inference time. This prefix replaces hundreds to thousands of natural-language Markdown skill tokens with a few virtual tokens, drastically reducing the input length. The researchers used Qwen3.5-4B as the backbone model. The soft skill is refined by a trainable soft delta while the base model remains unchanged. The paper also studies agentic execution as a harder boundary case, noting that "sparse trajectory imitation provides useful signal but does not yet robustly compress long-horizon procedural behavior."

Benchmark Results

SoftSkill was evaluated on three benchmarks in a single-round setting. The results show substantial improvements over both no-skill prompting and the existing SkillOpt method. The following table summarizes the accuracy gains:

Benchmark Improvement over No-Skill Improvement over SkillOpt
SearchQA +8.3 points +5.2 points
LiveMath +42.1 points +12.5 points
DocVQA +1.3 points N/A (not reported)

According to the paper, the improvement on LiveMath – a 42.1-point jump – is particularly notable, indicating that the compressed latent prior captures procedural math reasoning effectively. On SearchQA, SoftSkill achieved an 8.3-point improvement, and a smaller but positive 1.3-point gain on DocVQA. Relative to SkillOpt, SoftSkill improved accuracy by 5.2 points on SearchQA and 12.5 points on LiveMath, while using far fewer tokens.

Implications for Enterprise AI

For organizations deploying AI agents in domains like customer support, supply chain analytics, or knowledge base querying, the ability to compress behavioral policies into compact latent vectors could reduce inference cost and latency. The paper reports that SoftSkill "replaces hundreds to thousands of Markdown skill tokens with a few virtual tokens," directly addressing the token overhead problem. This approach may be especially relevant for enterprise environments where large numbers of specialized skills must be managed and updated frequently. However, the researchers caution that agentic execution – where an AI must follow long-horizon procedural behaviors – remains a challenge, and further work is needed to robustly compress such trajectories. The frozen-backbone design also means that the underlying model does not need to be retrained, simplifying deployment.

Overall, the results suggest that treating task skills as compact latent controls rather than reinterpreted Markdown text can yield significant performance gains while reducing computational demands. As the paper states, "some task skills are better treated not as additional Markdown to be reinterpreted at inference time, but as compact latent controls over how a frozen model enters the task."


Sources:

Keep Reading

Recommended Stories

Scientists Use AI and Quantum Computing to Generate New Peptides in Spare Time Technology

Scientists Use AI and Quantum Computing to Generate New Peptides in Spare Time

Researchers at the Technical University of Denmark used a hybrid AI-quantum computing system to generate novel peptides, achieving better results than classical models especially with limited data. The work, done on weekends with leftover funds, could accelerate personalized immunotherapies and vaccines.

July 12, 2026
New Research Shows Pretraining Data Composition Can Engineer Neural Scaling Laws for Particle Physics Technology

New Research Shows Pretraining Data Composition Can Engineer Neural Scaling Laws for Particle Physics

A new arXiv paper demonstrates that neural scaling laws in particle physics can be engineered by adjusting pretraining data composition. The study shows that including more diverse and task-aligned synthetic data can shift scaling behavior to require more data rather than larger models, offering insights for efficient AI training.

July 8, 2026
Researchers Propose Feature Selection to Improve Neural Additive Model Efficiency and Interpretability Technology

Researchers Propose Feature Selection to Improve Neural Additive Model Efficiency and Interpretability

A research paper proposes adding feature selection mechanisms to Neural Additive Models (NAM) and Neural Basis Models (NBM) to reduce computational costs and enable handling of feature interactions in high-dimensional datasets. The method updates selection weights during training, achieving better or comparable performance to state-of-the-art GAMs.

July 8, 2026
New Research Reveals How Visual Tokens Evolve Inside Vision-Language Models Technology

New Research Reveals How Visual Tokens Evolve Inside Vision-Language Models

A new computer vision paper from arXiv investigates how visual tokens are integrated into large language models (LLMs) under two paradigms: in-context prompting and layer-wise injection. The authors find that visual tokens enter the LLM as 'disguised visual context' lacking linguistic structure, then evolve differently depending on the integration architecture. They show that attention allocation alone is insufficient, and performance depends on the quality of visual representations at each layer.

July 8, 2026